SEM analysis of a cerium oxide polishing slurry reclamation process

Debra Saulnier, Department of Chemical Engineering and Laboratory for Laser Energetics

Introduction

Cerium oxide is a common polishing abrasive used in optical manufacturing. Recently, due to price instability and potential environmental impact, there is increased interest in reclaiming spent polishing slurry rather than discarding the used slurry as waste. A Gorham, NY company, Flint Creek Resources, Inc., has provided samples from various stages of their polishing slurry reclamation process for analysis.

Eight samples were analyzed in total, their names and descriptions are given in the table below.

Sample Name Description
Unicer 166 Fresh polishing abrasive
A001-1B Spent polishing abrasive
A001-1D Solids removed by 1st separation
A001-1C Solids recovered from 1st separation
A001-6A Solids removed by 2nd separation
A001-6B Solids recovered from 2nd separation
A001-1 Solids from waste
ECHO-SR6 Reprocessed polishing abrasive

 

Materials and Methods

The instrument used for this analysis is the Zeiss Auriga CrossBeam SEM-FIB  of  the Institute of Optics at the University of Rochester.  

Samples were imaged by normal methods with secondary and backscatter electrons. Additionally, EDS spectra were recorded. All samples were lightly milled with a mortar and pestle, put on sample stubs with carbon tape, then sputter-coated with gold at 10 mA for 150 seconds total to prevent charging. Some decoration artifacts are evident due to the long coating time.

Optical micrographs were taken on the Leica DMRX of the Laboratory for Laser Energetics at the University of Rochester.

Results

The resultant data consists of optical micrographs, secondary and backscatter electron images, and EDS spectra. All impressions are subjective and qualitative in nature; however, results are in good agreement with expectations.

Fresh and Spent Polishing Abrasive

Optical Micrographs
Unicer 166: Fresh A001-1B: Spent
Optical micrograph of Unicer 166 50x Optical micrograph of A001-1B 50x
Optical micrograph of Unicer166 100x Optical micrograph of A001-1B 100x
Secondary Electron Micrographs
Unicer 166 A001-1B
secondary electron micrograph of Unicer 166 1kx Secondary electron micrograph of A001-1B 1kx
Secondary electron micrograph of Unicer 166 5kx Secondary electron micrograph of A001-1B 5kx
Secondary electron micrograph of Unicer 166 10kx Secondary electron micrograph of A001-1B 10kx
Secondary electron micrograph of Unicer 166 20kx Secondary electron micrograph A001-1B 20kx
Backscatter Electron Micrographs
Unicer 166 A001-1B
Backscatter electron micrograph of Unicer166 1kx Backscatter electron micrograph of Unicer 166 1kx
Backscatter electron micrograph of Unicer 166 5kx Backscatter electron micrograph of A001-1B 5kx
Backscatter electron micrograph of Unicer 166 10kx Backscatter electron micrograph of A001-1B 10kx
EDS
Unicer 166 A001-1B
EDS spectrum for Unicer 166 EDS spectrum of A001-1B

 

The gold peak in the EDS spectrum is from sputter coating the sample. The primary contaminant in the spent slurry is silicon dioxide from polishing glass. Carbon contamination likely occurs due to degradation of polyurethane polishing pads; however, it is impossible to quatify with the sample mounted on carbon tape.

First Separation

Optical Micrographs
A001-1D: Removed A001-1C: Recovered
Optical micrograph of A001-1D 50x Optical micrograph of A001-1C 50x
Optical micrograph of A001-1D 100x Optical microgrph of A001-1C 100x
Secondary Electron Micrographs
A001-1D A001-1C
Secondary electron micrograph of A001-1D 1kx Secondary electron micrograph of A001-1C 1kx
Secondary electron micrograph of A001-1D 5kx Secondry electron micrograph of A001-1C 5k
Secondary electron micrograph of A001-1D 10kx Secondary electron micrograph of A001-1C 10kx
Secondary electron micrograph of A001-1D 20kx Secondary electron micrograph of A001-1C 20
Backscatter Electron Micrographs
A001-1D A001-1C
Backscatter electron micrograph of A001-1D 1kx Backscatter electron micrograph of A001-1C 1kx
Backscatter electron micrograph of A001-1D 10kx Backscatter electron micrograph of A001-1C 5kx
Backscatter electron micrograph of A001-1D 20kx Backscatter electron micrograph of A001-1C 10kx
EDS
A001-1D A001-1C
EDS spectrum for A001-1D EDS spectrum for A001-1C

 

The material removed in the first separation is primarily glass waste. Interestly, fibers (suspected to be polyurethane) are visible in the higher magnifaction backscatter micrographs. Due to the relatively low atomic number of the compositional elements, the fibers appear dark in the backscatter images.

Second Separation

Optical Micrographs
A001-6A: Removed A001-6B: Recovered
Optical micrograph of A001-6A 50x Optical microgaph of A001-6B 50x
Optical micrograph of A001-6A 100x Optical micrograph of A001-6B 100x
Secondary Electron Micrographs
A001-6A A001-6B
Secondary electron micrograph of A001-6A 1kx Secondary electron micrograph of A001-6B 1kx
Secondary electron micrograph of A001-6A 5kx Secondary electron micrograph of A001-6B 5kx
Secondary electron micrograph of A001-6A 10kx Secondary electron micrograph of A001-6B 10kx
Secondary electron micrograph of A001-6A 20kx Secondary electron micrograph of A001-6B 20kx
Backscatter Electron Micrographs
A001-6A A001-6B
Backscatter electron micrograph of A001-6A 1kx Backscatter electron micrograph of A001-6B 5kx
Backscatter electron micrograph of A001-6A 10kx Backscatter electron micrograph of A001-6B 10kx
Backscatter electron micrograph of A001-6A 10kx Backscatter electron image of A001-6B 20kx
EDS
A001-6A A001-6B
EDS spectrum for A001-6A EDS spectrum for A001-6B

 

The second separation removes most of the silicon and aluminum, but also some of the rare earth oxides that the polishing slurry is comprised of.

Waste and Reprocessed Polishing Abrasive

Optical Micrographs
A001-1: Waste ECHO-SR6: Reprocessed abrasive
Optical microgrph of A001-1 50x Optical micrograph of ECHO-SR6 50x
Optical micrograph of A001-1 Optical micrograph of ECHO-SR6 100x
Secondary Electron Micrographs
A001-1 ECHO-SR6
Secondary electron micrograph of A001-1 1kx Secondary electron micrograph of ECHO-SR6 1kx
Secondary electron micrograph of A001-1 5kx Secondary micrograph of ECHO-SR6 5kx
Secondary electron microgrph of A001-1 10kx Secondary electron micrograph of ECHO-SR6 10kx
Secondary electron micrograph of A001-1 20kx Secondary electron micrograph of ECHO-SR6 20kx
Backscatter Electron Micrographs
A001-1 ECHO-SR6
Backscatter electron micrograph of A001-1 1kx Backscatter electron micrograph of ECHO-SR6 1kx
Backscatter electron micrograph of A001-1 5kx Backscatter electron micrograph of ECHO-SR6 5kx
Backscatter electron micrograph of A001-1 10kx Backscatter electron micrograph of ECHO-SR6 10kx
EDS
A001-1 ECHO-SR6
EDS spectrum for A001-1 EDS spectrum for ECHO-SR6

 

The waste material is primarily silicon dioxide with some aluminum contamination. The presence of nickel in this sample seems abnormally high since it does not appear in solids removed in the first of second separation; a nickel-rich area may have been selected accidentally when performing EDS analysis. Again, the strong carbon peak in the waste might indicate the presence of polyurethane in the waste, but is likely also (perhaps moreso) due to the carbon tape used to mount the sample. Looking at particle size, and the electron flight simulations given below, it is apparent that the electron beam sees the carbon tape beneath small silica particles. A lower accelerating voltage could have been used.

Conclusions

SEM/EDS analysis revealed that the composition of the reprocessed polishing abrasive (ECHO-SR6) in nominally the same as the fresh polishing abrasive (Unicer 166), indicating that rare earth oxide polishing abrasives can be reclaimed from spent slurry.

Unicer 166 ECHO-SR6
EDS spectrum for Unicer 166 EDS spectrum for ECHO-SR6

The average particle size qualitatively decreases through use and reclamation; the result is a final product that is finer than the starting material.

It would be useful to prepare and analyze a second set of samples to confirm that there is no sampling bias based on sample preparation methods. Sample vials were agitated before preparation, but samples did not appear to be homogeneous across the sample stub; the images presented represent the most typical regions of the sample.

Acknowledgements

I would like to thank Mark Mayton from Flint Creek Resources, Inc. for providing the samples, Brian McIntyre and Rohit Puranik  of University of Rochester’s Institute of Optics for their guidance regarding the imaging of these samples, and Stephen Jacobs and Tess Jacobs of the Laboratory for Laser Energetics for their support throughout this process.

Student Aid Online


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

Your name:

Email address: