Christian Freitag and Roswitha Giedl-Wagner
X-Ray Protection in an Industrial Production Environment
Industrial implementation of the requirements of the German Radiation Protection Act
With the classification of ultrashort-pulse laser processing systems as “systems for the generation of ionizing radiation” in the German Strahlenschutzgesetz (Radiation Protection Act), their operation is usually no longer possible without a license and without notification of the authorities. This article deals with the implementation of the legal requirements resulting from this classification in an industrial production environment.
Using ultrashort-pulse lasers for materials processing offers many advantages compared to laser sources with longer pulse durations: minimum heat input, a mechanical precision in the micro-meter range and the ability to process transparent materials are just a few of them . The typical pulse duration of industrial laser sources used for materials processing is in the range of 300 fs to 10 ps. Using such short pulse durations, an irradiance of more than 1013 W/cm² is easily achieved. With such high laser irradiances, the electron temperatures in the generated plasma reach several keV which results in bremsstrahlung with emissions in the keV X-ray order. This emission of X-ray radiation when ablating material with ultrashort-pulse lasers was already investigated in the 1980s . Although it is an already known effect, it has recently come increasingly into focus with the increase in the average power of ultrashort-pulse lasers into the kW range. Since the dose rate increases proportionally to the average power, X-ray emission from processes using higher average laser powers can represent a serious health risk. Accordingly, research activities in this area have increased strongly in recent years. Legall et al. (2018) measured the spectral X-ray emission for different target materials using an ultra-short-pulse laser with a wavelength of 1030 nm. X-ray photons with energies up to 25 keV were detected . These results were confirmed by Behrens et al. (2019) . Further research was carried out on the influence of processing parameters on X-ray emission by Legall et al. (2019) . It was found that the intra-line separation and the scanning direction have an influence on the measured X-ray dose when using a beam scanning scheme. An analytical model to estimate the expected X-ray dose when laser processing with ultrashort- pulse lasers under industrial conditions was presented by Weber et al. (2019) . Using the presented model, it was concluded that shielding with a 2 mm sheet of iron is, to current knowledge, sufficient to attenuate the radiation level generated during processing with a 1 kW ultrashort-pulse laser to a safe level. This prediction was recently confirmed by Legall et al. (2020) . The potential health risk posed by the X-rays generated during material processing with ultrashort-pulse lasers led to a reaction from the legislator. The new German Radiation Protection Regulation came into force on December 31, 2018, which also applies to material processing stations equipped with an ultrashort-pulse laser. This article deals with the effects of this regulation on a German company.
Part C of the German Radiation Protection Regulation states: “…The operation of systems for the generation of ionizing radiation in which ionizing radiation can be generated by the incidence of laser radiation according to § 2 paragraph 3 sentence 1 of the Regulations on Artificial Optical Radiation on material is exempt from licensing and notification if the irradiance of the laser radiation does not exceed 1 ⋅ 1013 watts per square centimeter and the local dose rate in 0.1 meter from the touchable surface does not exceed 1 micro Sievert per hour.” One of the consequences of this regulation is that companies have to acquire a license for the operation of every processing station that is operated with an irradiance exceeding 1 ⋅ 1013 W/cm². This level is easily possible using an ultrashort-pulse laser. An application must be submitted to the authorities to obtain this license and should include:
- The appointment of a radiation protection officer and a radiation protection commissioner who have the necessary technical qualifications
- Documentation including, amongst other things, the annual maintenance plan, planned materials to be processed, a safety report and a hazard assessment
- Proof of coverage against radiation damage provided by the insurance company
- Radiation protection training and a list of the persons working with the system
- A test report according to the measurement specifications
- Documentation of the continuous dosimetry
A critical point is the test report according to the measurement specifications, where an X-ray test qualifying the housing must be performed. The aim is to guarantee the X-ray safety of the housing so that there is no danger to the operator. Of special interest are the doses H(0.07) and H(10) and the corresponding dose rates H. (0.07) and H. (10). These describe the absorbed X-ray energy per mass at a depth in the body of 0.07 mm and 10 mm, respectively. The first value mainly considers X-ray photon energies around 5 keV; the latter photon energies at about 30 keV . The corresponding dose rates consider the absorbed X-ray energy per mass and time. The legal limit for H(0.07) is an accumulated dose of 50 mSv within one year and for H(10) an accumulated dose of 1 mSv within one year. The legal requirement that the local dose rate at 0.1 meter from the touchable surface should not exceed 1 μS/h assumes the H. (10) dose rate. In the following, an X-ray test of a housing is presented as an example of a real application case.
The tested laser processing station was a GL.compact from GFH. This station is equipped with an ultrashort-pulse laser with a pulse duration of 900 fs, a wavelength of 515 nm, a maximum average power of 35 W, a repetition rate of 200 kHz, a maximum pulse energy of 174 μJ and a focal spot diameter of 10 μm. This results in a maximum intensity at the focus of 4.7 ⋅ 1014 W/cm². A scanner system was used for beam movement since, according to current knowledge, hatching causes the highest X-ray dose. The laser generated plasma is shielded least using a hatched scanning process. An overview of the experimental setup is shown in Fig. 1. The beam is focused on the workpiece with focusing optics. Two DIS-1 dosimeters from Mirion Technologies were placed inside the station near the processing zone, one at a distance to the processing zone of 75 mm and one within a distance of 200 mm. A sketch of the setup and the processing strategy is shown in Fig. 2. In addition to the dosimeters placed inside the stations, dosimeters were placed outside at the critical points of the housing. One dosimeter was placed at the laser safety window and one dosimeter was placed on the housing at the shortest distance to the processing zone. The minimum distance from the processing zone to the housing is 300 mm. Another dosimeter is placed at a transition between the housing and a loading door, as this is a potential weak point. The housing itself is made out of 2 mm thick steel. The laser security window is 3 mm thick, additionally secured by a lead glass pane with a lead equivalent value of 1.5 mm. A protruding flange is installed at the openings, for example at the transition from the housing to the door, so that a direct view of the laser plasma is always blocked by at least 2 mm steel.
As a preparation for the housing test, the generated X-ray dose was measured for two different materials, steel and tungsten, since both materials showed similar dose rates in previous studies . The dose rate was measured as a function of the irradiance. The irradiance was varied by changing the pulse energy. The average laser power was kept constant at 3.5 W by changing the repetition rate accordingly. A constant average laser power is important, since the dose rate increases linearly with the average laser power . To keep the pulse overlap constant, the movement speed of the laser beam was changed according to the repetition rate. The dose rate was measured using an OD-02 dosimeter from STEP Sensortechnik and Elektronik Pockau. It was placed inside the processing station at a distance of 200 mm from the processing zone. The measured dose rate as a function of the irradiance can be seen in Fig. 3. The dose rate is about a factor of ten higher when processing tungsten compared to steel. For this reason, tungsten was used for the subsequent housing test. It can also be seen in Fig. 3 that there is a linear correlation between the dose rate and the irradiance. The dose rate is maximum at maximum irradiance, and the maximum irradiance was used for the housing test.
Previous studies showed that the scanning direction has a significant influence on the measured X-ray dose . As a further measure in preparation for the housing test, the process was examined for the dependence of X-ray emission on the hatching angle. The dose rate was again measured using the OD-02 dosimeter during a scanning process with eight different orientations of the hatching pattern. The dosimeter was again placed inside the processing station at a distance of 200 mm from the processing zone. Squares with a size of 28 × 28 mm were ablated on tungsten. The orientation of the hatching pattern was rotated by 45° after each processing step. The total process time was 250 s, so the hatch pattern was rotated approximately every 30 s. The measured dose rate as a function of the processing time can be seen in Fig. 4. The individual process steps are separated and numbered by red dotted lines. There is a strong dependence of the dose rate measured on the hatching direction with respect to the dosimeter. With optimum alignment of the hatching direction, up to five times more X-rays are recorded than with the most unfavourable alignment. Furthermore, it can be seen that the maximum dose was not generated during the first pass but only during the eighth process step. There the alignment of the dosimeter to the hatching direction seemed to be optimum. The positioning and hatching direction during step 8 are depicted in Fig. 2. This strategy of rotating the hatching direction in 45-degree steps was used for the housing test. Such a rotation pattern ensures that all dosimeters (inside and outside the station) are exposed to the same radiation dose over time. In this way a comparable measurement result can be achieved.
X-ray test of the housing
The maximum average laser power of 35 W was used for the housing test. The total duration of the measurement was 96 minutes. The dosimeters were read out every 16 minutes. Please note that the dosimeters were not reset after each read out. In this way, the temporal development of the dose can be tracked. The measurement results for the two dosimeters located inside the station are shown in Fig. 5. In Fig. 5a the accumulated dose H(0.07) is shown, in Fig. 5b the accumulated dose H(10) is shown. A high H(0.07) dose was measured at a distance of 75 mm to the processing zone. After one hour, a dose of 102 mSv was achieved. The legal limit is 50 mSv, which is already exceeded close to the processing zone and without shielding after half an hour. Increasing the distance to 200 mm (again without shielding), the H(0.07) dose already drops significantly. After one hour, a dose of about 2 mSv was measured. These measurements confirm that keeping a sufficiently large distance to the processing zone already significantly reduces the dose. The accumulated dose of H(10) at a distance of 75 mm was 1 μSv after 16 minutes and stayed constant afterwards. At a distance of 200 mm the H(10) dose was 0 μSv for the whole duration of the measurement. A reliable measurement is difficult at such low values because the readout accuracy of the H(10) dose of the DIS-1 is 1 μSv. As the accumulated dose of H(10) at 75 mm distance did not increase over the prolonged exposure it can be assumed that the readout of 1μSv is not representative for the real emission but caused by the readout resolution. The insufficient readout accuracy is problematic in that the legal limit is 1 μSv per hour. However, since at a distance of 200 mm the measured H(10) dose was 0 μSv, it can be expected that outside the housing, which has a minimum distance to the processing zone of 300 mm, no X-ray emission relevant for the H(10) dose will be present. Three dosimeters were placed at critical points on the outside of the housing (window, housing with the shortest distance to the processing zone, transition between housing and door). All of these dosimeters measured a H(0.07) dose of 0 mSv and a H(10) dose of 0 μSv. Although, as shown before, inside the housing, a significant amount of H(0.07) dose was measured, no X-ray radiation could be detected outside the housing. The housing therefore provides sufficient shielding against the X-ray radiation generated during laser processing with an ultrashort-pulse laser.
The need to apply for a licence as well as the conditions required for the granting of the licence cause additional costs in a company. For the license application, the preparation of a test report, the attendance of a radiation protection course to act as radiation protection officer as well as the issuing of the license, a one-time cost of approximately 6,000 euros is incurred. The radiation protection course has to be revisited every five years. This is a cost estimate based on the current situation (March 2020) and the examination of the radiation-proof designed GL.compact. The costs in individual cases may vary considerably. If the testing process is further standardized, the time for the housing test and therefore the costs may be reduced. The testing process may be more complex and thus more cost-intensive for other, more complex systems than the GL.compact. Additional costs of about 650 euros per year arise for the continuous dosimetry as well as the insurance coverage. In addition, the resulting administrative costs for the radiation protection organization must be taken into account. All in all, the necessary administrative effort and costs for companies should not be underestimated when commissioning an ultrashort-pulse laser processing station.
Since the entry into force of the German Radiation Protection Regulation, a licence is now required for the operation of ultrashort-pulse laser processing systems. This requires a report on the radiographic test of the housing. Such a test was described in an actual application example and the difficulties illustrated with the current method. The housing consisted of 2 mm of steel. Protruding flanges were implemented at potential weak points, like the opening between a door and the housing. The laser safety window was additionally secured by lead glass. These security measures were sufficient, so that no X-ray emission was measured outside the housing even though a significant amount of X-ray emission was generated during ultrashort- pulse laser processing.
The authors would like to thank Dr. Rudolf Weber from the Institut für Strahlwerkzeuge (IFSW) of the University of Stuttgart for his support during the measurements.
 S. Nolte: Ultrashort Pulse Laser Technology, Laser Sources and Applications, Springer Series in Optical Sciences Volume 195, Springer International Publishing Switzerland 2016, ISBN: 978-3-319-17658-1,  Weber et al.: Appl. Phys. Lett. 53 (1988) 2596,  Legall et al.: Applied Physics A 124 (2018) 407,  Behrens et al.: Radiation Protection Dosimetry 183 (2019) 3, 361-374  Legall et al.: Applied Physics A 125 (2019) 570,  Weber et al.: Applied Physics A 125 (2019) 635,  Legall et al.: J. Laser Appl. 32 (2020) 022004
Source all images: LightPulse Laser Precision