Ascertaining the temperature distribution on quartz glass surfaces during the laser polishing process.

Hecht, Kerstin ; Bliedtner, Jens ; Muller, Hartmut 等

Abstract: Polish optical or technical glass surfaces makes high demands on polishing agents, polishing compound carriers and the technician himself. Especially micro-geometries and freeform surfaces confine the use of traditional abrasives and tools. The laser beam as a polishing tool is able to solve the problems and to polish each surface and outline or geometry on it in the shortest possible time.

The problem of the laser polishing process is the temperature distribution developing on the surface. Because of the punctual application of laser energy the glass part is heated irregular, which results in undesirable material tensions.

Pyrometric measurements detect the surface temperature and by the help of simulations it is possible to draw conclusions regarding the tensions in the glass.

Key words: laser polishing, quartz glass, temperature measurement

1. INTRODUCTION

Polishing is used generally to reduce the surface roughness to a value smaller than the wavelength of visible light (<400nm) (Grunwald, 1985). The necessary polishing process requires a lot of time, because it's carried out in several steps and a pre-editing (e.g. by lapping) is necessary. Furthermore the classical polish is bound to tools, which were not versatile for different surface geometries. Some times it is even necessary to finish optical or technical surfaces manual. This means a time--and cost-intensive production of the given glass components. With the laser beam you'll have a rotation-symmetrically shaped tool with an extremely small diameter that is able to move in a three-dimensional space if a corresponding controlling unit is used. The achievable average polishing time is within seconds (ILT, 2003).

The laser polishing process of quartz glass is, as far as possible, a method without material removal, where a defocused beam melts the glass surface. During the solidification the surface tension in the melted layer is leading to the smoothing of the roughness profile. The surface quality, you can reach with laser polishing, depends on different parameters. A great importance is attached to the surface temperature. This one, caused by laser output power, feed-rate and beam velocity, should be measured contactless. To do this a pyrometer is used.

2. EXPERIMENTAL PROCEDURE, RESULTS

The infrared radiation (also known as thermal radiation) was detected in 1666 by Sir Isaac Newton, as he dispersed light with a prism (OMEGA NEWPORT, 2007). Each body with a temperature above the absolute zero (0 Kelvin) emits an electromagnetic radiation from its surface, which is proportional to its intrinsic temperature. A part of this radiation (also called intrinsic radiation) is infrared radiation, which can be used to measure the temperature of a body. This radiation penetrates the atmosphere. With the help of input optics the beam is focused on a detector element, which generates an electrical signal (voltage U), proportional to the radiation. The signal is amplified and transformed into an output signal, proportional to the object temperature ([T.sub.object]) (Optris, 2006).

According to the Stefan-Boltzmann law is U ~ [T.sub.body] x [epsilon] whereas the emission ratio [epsilon] depends on several factors (material, surface character, temperature, wave length, measurement setup, etc). It was not possible to find out a satisfactory declaration of value of [epsilon] for the available quartz glass in literature. That's why a black body was used in an experiment to determine the correct value.

The applied pyrometer works at a wave length of 5.14[micro]m. Quartz glass is non transparent for this wave length, that means there is no transmittance of radiation, just the surface temperature will be detected.

Figure 1 shows the machining and measurement setup. The [co.sub.2] laser (wave length 10.6[micro]m) is diverted in lines by a mirror and meets the glass surface defocused. The feed is realized by a driven axis whereas just the glass sample moves relatively to the rest of the system. The pyrometer is aligned in a way that the spot size (measured area) with a diameter of 1.4mm is located right behind the laser beam line.

By means of comprehensive experiments the influence of various parameters on the surface temperature should be detected. So the following parameters were changed:

* laser output power P (500 ... 690W)

* laser beam velocity vs (400 ... 1000mm/s)

* feed rate in 3 sections

* length of the sections

* beam guide concept

The temperature is measured always at the same spot right behind the laser beam (cp. Fig. 1, detail).

The samples out of quartz glass are 6mm thick and 25mm squared. Their surfaces were pre-edited differently (grinded) in groups to generate diverse roughness classes. After the temperature measurement also the influence of the different parameters on the reachable surface qualities should be checked.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

Now the results of the experiments should be displayed in some examples.

Temperature and laser output power behave nearly proportional, with declining power (increased application of energy) temperature is rising.

Laser beam velocity and temperature behave inversely proportional in principle (Fig. 2). The beam is guided faster over the surface, keeping the feed rate constant and the temperature is falling. Because of this that the distance between the polishing lines is getting smaller in this trial, you could probably expect a higher application of energy per area and a higher temperature in suggestion. On the other hand the application of energy had to happen in a shorter time, so the temperature stays low. Another unusual aspect is the jump between 600 and 800mm/s. The laser needs obviously a certain time to apply its energy into the surface layer. It's expected that rougher surfaces have to be polished with lower beam velocities, so that the radiation is possible to penetrate the surface deep enough to melt not only the profile peaks but the whole roughness profile.

A basic criterion for uniform polishing without or with low stresses is the temperature-time-behaviour. The glass part can be divided in 3 sections. In the 1st one ([s.sub.1]) the part is warmed-up to polishing temperature. This has to go quickly, so that the glass part is polished right from the beginning.

The 2nd sector ([s.sub.2]) should be characterised by a constant temperature without any rise or fall. In the 3rd section there is, without intervention, a massive increase in temperature because there is a heat accumulation at the end of the sample. That will lead to stresses at the edge and should be avoided.

Figure 3 shows the consequences of three different position-velocity-systems.

[FIGURE 3 OMITTED]

The sections ([s.sub.n]) are given in mm and the velocities in m/min. At the change from the 1st to the 2nd system the velocity is reduced while section [s.sub.1] is shortened. The result is that there is a sharp increase now and the cant of temperature at the beginning is reduced. Next it's temped accessorily to avoid the heat accumulation at the end of the part. This is possible by increasing [v.sub.3] and shortening [s.sub.3]. At the same time [v.sub.2] is reduced so that the cant of temperature at the beginning disappears completely. It's impossible to avoid the temperature peak at the end of the quartz glass part at all.

By the reason that the polishing process passes not entirely without material removal sublimate could deposit on the already polished surface. Just with changing laser parameters this is not preventable. So the beam guide concept had to be adapted. Figure 4 shows (a) a beam guided simply in meandering courses with constant velocity, which leaves sublimate clearly. In case (b) it is recognizable that, because of the beam, which runs after itself with a high velocity (2000mm/s), a surface free of sublimate is left behind.

Measures of roughness show the improvements of surface quality impressively. In one example it was possible to reduce Ra from 460nm to 3.4nm. This was achieved at the end of one experimental series where the parameters have been optimized.

[FIGURE 4 OMITTED]

3. CONCLUSION AND PERSPECTIVES

During the experiments. Lasting several days, the pyrometers positioning (and the position of the measure spot respectively) turned out as a problem. Despite of a very stable assembling of the measurement device to the laser system the position has changed minimal because of the reference run. This has, indeed, just a small effect (fraction of a millimetre) on the sample surface but in the measured temperature this causes a temperature difference of e.g. 250 [degrees]C in one and the same trail.

The results of the temperature measurement, in connect the fallowing measure of roughness, should be lead to an automatized laser polishing process. To do this stable fixing of the pyrometer, analysis or simulations of material tensions and a laser controlling unit (based on temperature) are required. The automatized laser polishing process is meant to be u customer-specific tasks in polishing and should be applied in all areas of using and processing glass.

The method was already tested to produce quarts glass mould inserts for injection moulding tools. These inserts assure high durability of the moulds, guarantee a higher than average abrasion resistance and behave indifferent against the used plastics.

4. REFERENCES

Fraunhofer Institut fur Lasterchnik, ILT; (2003). Fraunhofer ILT Jahresbericht (Fraunhofer ILT Annual Report 2003) 2003, Available from: http://publica.fraunhofer.de/eprints/N-35932.pdf, Accessed: 2007-08-06

Grunwald, F.; (1985). Fertigunsverfahren in der Geratetechnik (Manufacturing Methods in Apparatus Technology), Carl Hanser Verlag, 3-446-14195-2, GDR

Newport Electronics GmbH (2007). Einfuhrung in Infrarot-Pyrometer (Introduction to Infrared Pyrometers), Available from: www.omega.de Accessed: 2007-07-23

Optris GmbH (2006). Basics of Noncontact Infrared Temperature Measurement, Available from: www.optris.de/en Accessed: 2007-07-23.