A Microscopic Analysis Of Conventional Optical Polishing Materials

Tim Lynch

Department of Mechanical Engineering
University of Rochester
Rochester, NY


Introduction

The practice of polishing glass and other materials for use as optical components dates has been around for many centuries, dating back to the time of Isaac Newton. Throughout the years, numerous polishing materials have come in and out of favor; however they all operate under the same basic principles. In order to remove material and create the desired optical grade surface, an abrasive particle is brought into contact with the optical substrate. These abrasive particles are either embedded in some sort of carrier material or held in suspension within a fluid solution. In the case of the embedded fixed abrasive, a non-abrasive pad is affixed to a tool in order to facilitate the interaction between the abrasive and the substrate.

The choice of carrier pad is crucial in the effective polishing of an optical substrate. Material properties specific to a particular substrate dictate the necessary characteristics in the pad and slurry combination. To better understand the properties of these polishing pads, a number of pad material samples were examined using a scanning electron microscope. The goal was to gain an appreciation of the pad characteristics at a microscopic level in order to facilitate more informed decision making when choosing a polishing pad material.


Methods

Imaging for this lab was performed on a Zeiss Auriga Scanning Electron Microscope housed in the Wilmot building on the University of Rochester campus. To make this study applicable to a wider array of polishing scenarios, five unique polishing pads were chosen for examination in the microscope:

  1. Coarse grit embedded diamond
  2. Fine grit embedded diamond
  3. Polyurethane synthetic
  4. Polyester felt
  5. Polyester felt (not used)
  6. Closed cell foam

These pads are representative of the basic categories of pads found in most optics shops, allowing for conclusions drawn from a particular sample to be applied other similar pads of the same type. Further description of each pad is found below in the respective results section.

In order to sufficiently characterize the pads under test, a number of techniques were performed as part of the study. These techniques include but are not limited to:

  1. Gold Coating of Non-Conductive Samples
  2. Electron Trajectory Analysis
  3. Secondary Electron Detection
  4. Back Scatter Electron Detection
  5. X-ray Elemental Analysis
  6. Colorization of Acquired Images

Details of these processes will also be discussed in further detail where applicable as part of the results section. All sample preparation was carried out in the Wilmot building laboratories.


Results

Results from the study will be discussed within the context of each polishing pad material. All images from the SEM were acquired utilizing the secondary electron detector unless otherwise noted.

Coarse Grit Embedded Diamond

As the name implies, embedded diamond pads consist of diamond particles adhered to a backing material. These pads can be thought of as a sort of optical grade sandpaper and are used when large quantities of material removal is desired.

Examination of the acquired SE2 images shown at left reveals a somewhat random and non-uniform distribution of abrasive across the surface of the pad. This revelation was somewhat unexpected but begins to make sense after further thought. It is desirable to have uniform removal from these pads across the optical surface, which one would most commonly associate with uniform particulate distribution; however it is also important to consider the mode in which these pads are used. Most common grinding and polishing set ups feature the pad spinning at a substantial velocity relative to the optical surface. The rapid movement of the pad across the surface incorporates a random element into the pad and surface interaction, offsetting the uneven abrasive distribution. This combination of effects ultimately creates the desired uniform surface removal.

To predict the interaction volume of the electrons through the embedded diamond, an analysis was performed in Win X-Ray. Win X-Ray is a program that uses a Monte Carlo analysis to approximate electron trajectories through a particular material. Since pure diamond consists entirely of carbon atoms, the analysis used carbon as the substrate material. All other parameters were matched to the values used in the SEM imaging.

The results of the simulation are presented as a visual mapping of the predicted electron paths. Observation of the path mappings reveals that much of the electron propagation is straight down into the specimen with relatively little scatter. This results in a long, narrow interaction volume that pierces deeper into the sample.

In an effort to further understand the makeup of the embedded diamond pad, an elemental analysis of the specimen was performed in addition to the imaging discussed above and shown in the top left image of the above figure. Two processes were used to obtain this compositional analysis: back scatter detection and x-ray analysis. Back scatter detectors capture electrons deflected by the atomic forces inside each element, thereby creating contrast in the image due to the different relative forces of each element. These images highlight high and low atomic masses through light and dark areas in the image. The bottom left image in the above figure shows the results from the BSD. The diamond particles are quite dark due to their high atomic mass, while the back material appears much lighter. This information was used as a baseline for comparison when examining the data from the x-ray analysis.

Additional compositional information was obtained through x-ray analysis. This process utilized the EDAX x-ray spectrometer system, an additional component integrated into the SEM. In addition to an x-ray detector, the EDAX system has mapping capabilities which can return information regarding the location of different elements within the specimen. Locations of carbon (purple), oxygen (yellow), and nickel (green) are shown in the top three images. Analysis of the carbon map reveals that location of carbon deposits align with those of the diamond particles as mapped by the SE2 and BSD images. Give that diamonds are primarily carbon, this provides a strong endorsement of the accuracy of the mapping.

The EDAX system can also acquire information pertaining to the relative concentrations of each element, shown in the bottom left spectrum. Not surprisingly, carbon is the predominant element, followed by oxygen and nickel. Gold also appears in the spectrum due to the gold coating on the specimen.

Fine Grit Embedded Diamond

The fine grit embedded diamond pad consists of the same materials as the coarse grit pad but uses a smaller diamond particle. The smaller abrasive leads to less agressive material removal and would traditionally be used as a second step after processing with the coarser pad.

The image to the left was taken at the same magnification as the first image of the coarse diamond pad, allowing for direct comparison between the two materials. As expected, there is a clear distinction between the size of the diamond in each image. There was particular interest as to whether the finer grit pad would feature a more uniform distribution of diamond particles, but this does not appear to be the case. As discussed above, the mode in which the pad is used potentially compensates for this non-uniformity and results in the desired uniform material removal across the optical surface.

Polyurethane Synthetic

As opposed to the two previously discussed diamond pads, the polyurethane synthetic pad does not rely solely upon abrasive embedded in the material. Rather, the pad is used as a carrier to move loose abrasive across the optical surface being polished. This abrasive is most commonly suspended in solution which floods the part, providing ample opportunity for the pad to temporarily envelop abrasive particles and force them into contact with the optic.

An examination of the polyurethane synthetic pad's surface topography reveals why this type of polishing action works so well. The upper, purple in lens image shows the numerous cavities covering the pad's surface. Formed by injecting additional additives into the mixture during manufacturing process that later dissolve, the pad maThese cavities provide ample space in which to capture the abrasive slurry solution and pull it across the optical surface.

In the case of this particular polyurethane synthetic pad, the abrasive slurry solution is not the only source of material removal. Closer examination of the lower, teal image reveals small particles scattered across the pad surface. These are small abrasive particles embedded in the pad which add in mechanical removal. The combination of loose and embedded abrasive makes tools such as these particularly useful for bulk material removal.

Polyester Felt

Polyester felt is another pad used as a carrier during the polishing process. Similar to the polyurethane synthetic material, abrasive is trapped in the material and dragged across the optical surface. Pads such as this come in varying thickness to control the amount of pressure applied to the surface being polished.

Closer examination of the polyester felt using the secondary electron detector provides insight as to how the pad carries abrasive particles. The material is made up of many tightly packed fibers, creating a dense mesh that captures abrasive. The relatively soft nature of these fibers make this pad an ideal choice for a final finishing step, as there is significantly less risk of scratching the optic with a sharp edge.

Closed Cell Foam

Though not used in quite the same way as the previously discussed pads, closed cell foam still plays a valuable role in optical manufacturing. Many aspects of the manufacturing process require some degree of compliance to avoid damage to either the part or the tool. Foam such as the one shown to the right can provide this compliance while still maintaining the integrity of the tool.

Foams gain their compliance from a network of pockets within the material. These pockets provide space for the material to compress into under an applied load, giving foam its characteristic springy nature. By varying the base material of the foam and the size of the pockets, manufacturers can control the behavior of the foam material.


Conclusions

Examination of a variety of materials used in optical manufacturing revealed a number of interesting relations between their microscopic structure and use in the optical manufacturing process. Materials with embedded abrasive possess a non-uniform distribution of abrasive particles but compensate through the relative rapid motion of the tool on which they are used. Materials utilized in conjunction with abrasives suspended in solution feature a numerous cavities that trap slurry that trap abrasive and drag it across the optical surface. Foams have a variety of pocket shapes and sizes that influence its compressibility. Ultimately, manufacturing a quality optical component is dependent on understanding the properties of each material and choosing pads that work best for the task at hand.

Acknowledgements

Many thanks to Brian McIntyre for his guidance throughout this project, and to Emily Hessney for her feedback and support.


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

Your name:

Email address: