Acoustic emissions – live monitoring of signals from the polishing gap

. While acoustic emissions are a method for e. g. the monitoring of bridges or for the assessment of the integrity of tanks, we wanted to investigate them for a further field of application. We tested AEs for the monitoring and assessment of a high-quality finishing process for optical components, the polishing of lenses. This process is - even though - it is of incalculable importance for nearly every optical component still a process which is commonly considered as some kind of black magic. With our research, we want to deliver an insight in this process with interacting mechanical and chemical processes. We want to present a method for the in-situ evaluation of a polishing process.


Glass polishing
The development of modern semiconductors requires high-quality optical systems.The final step of the optical fabrication process chain is the polishing step a , which is a process with mechanical and chemical parameters.These parameters interact in a complicated manner, depending on glass material, process parameters, polishing material, polishing carrier material, and ambient influences.Parameters are among others the relative velocity vrel and the pressure p, but also the hardness of the glass, the hardness, and type of the polishing carrier, chemical properties of the polishing slurry (e. g. pH-value, conductivity, etc.), as well as many other parameters [1].A complete list of process parameters would be exhaustive.
Through the nature of the glass polishing process -a process with rotating components in a wet and abrasive environment, it is difficult to get an insight into the process.Optical methods are efficient but fail at a rough interface (caused by the grinding process).The polishing gap is very small, it is not possible to integrate complete systems.

Acoustic emissions
Therefore, we suggest a method that allows the measurement of signals from the polishing gap, while the a If e. g. the coating and the cleaning is neglected.
sensory components are at a sufficiently large distance.The method we propose is the measurement of acoustic emissions (AEs).These are signals occurring during the generation of transient waves in a solid.This happens, e. g. during cracking processes and mechanical deformation, but also during corrosion, friction, and magnetic processes.Especially, cracking processes occur during polishing, as well as corrosion processes.This is our motivation to measure these signals during the polishing of optical components.The signals, we want to measure are AEs in the range of 100…500 kHz [2].

Experiments
For the evaluation of the application of AEs on glass polishing, we carried out several experiments with varying pressure, the relative velocity, the pressure, the glass type or the polishing carrier.The polishing carriers we used are shown in Table 1.

LP-87 47
The used polishing carriers are standard types which are nowadays common in optical polishing.They are of polyurethane type.

Results
The polishing was carried out with different polishing carriers and for different polishing situations.We were able to show, that it is possible to measure AEs during the polishing process.This is shown exemplarily in Fig. 1 for a not completely polished glass sample.However, it is possible to measure the AEs and see a dynamic in the evolution of the signals.We also carried out several experiments which serve the purpose to investigate and to understand the evolution of the signal and to interpret the dynamics during signal evolution.
For this purpose, we introduced a glass sample in the system and polished it for a longer time (25 minutes) with the removal of the sample every five minutes.This serves the purpose to measure the roughness of the sample.Fig. 2 shows the result of this measurement.The measured values show that the removal exits during a certain time.A low RMS value of the samples corresponds -by trend -with low roughness values.However, this must be investigated in detail and must be coupled with further sensor values (e. g. machine power).

Conclusion
We demonstrated that it is possible to measure the AEs during a process.Furthermore, we were able to couple by trend roughness values and AEs.The signal processing requires additional experiments and the process interdependencies have to be considered.Nevertheless, we assume that AEs are a powerful tool to get a live insight into the polishing process and to gather signals from the polishing gap itself.Therefore, we see a potential for this technology in the use of monitoring purposes for glass polishing, especially if coupled with additional sensors.

Fig. 1 .
Fig. 1.Polished glass sample for a non-completely polished glass surface.The inclompete polishing is already visible in the optical microscope.

Fig. 2 .
Fig. 2. AEs during the polishing of a glass sample with a LP-13 polishing carrier.The process parameters are: vrel: 1250 min -1 , p: 1 Bar.The glass type is N-BK7, a standard glass for optical components.Small steps between the individual measurements can be explained by the removal of the sample and by the restoration of a stable polishing situation.

Fig. 3 .
Fig. 3. RMS value of the AEs and average roughness of the glass sample.Due to the calculation of the RMS AE value in a certain time interval, the value at e. g. 5 minutes represents the mean RMS value in the time interval 0…5 minutes.