Advertisement

Nanomanufacturing and Metrology

, Volume 1, Issue 4, pp 260–267 | Cite as

Cleaning State of the Loop Case for Optical Crystal Module in Final Optics Assembly

  • Qingshun Bai
  • Yuhai Li
  • Kai Zhang
  • Canbin Wang
  • Xiaodong Yuan
  • Feihu Zhang
Original Articles
  • 46 Downloads

Abstract

In order to solve the online clean maintenance problem of the optical frequency-doubling crystal module in the final optics assembly, a crystal loop case is developed for better controlling the clean maintenance process. The study presents a detailed research on the cleanliness of the crystal loop case during the online maintenance process and the installation sequence of the crystal modules in the crystal. It first established a fluid simulation model of the crystal loop case system according to the online maintenance process and then analyzed the isolation effect of inlet velocity on the polluted air in view of the cleaning effect of the crystal loop case. Based on the simulation principle of gas–solid two-phase flow and the tracking results of the solid particle contamination, the range of air inflow is given for achieving the best cleaning status of the loop case. The optimal sequence of crystal module is explicit for its installation in the crystal loop case of frequency-doubling crystal module. The experimental setup has been built to examine the cleaning state of the crystal loop case, and the simulation result has been validated. The research on the cleanliness of crystal loop case can provide a useful reference for the ultra-clean manufacturing of line replaceable units and the closed loop control of cleaning in high-power laser facility.

Keywords

Loop case Cleaning state Particle contamination Final optics assembly Ultra-clean manufacturing 

1 Introduction

Inertial confinement fusion (ICF) as an effective method of achieving controlled fusion is developed as a strategic high technology all over the world [1]. National Ignition Facility in the USA, Laser Mégajoule in France, ShenGuang (SG) facility in China and others have been established over the years. The final optics assembly (FOA) is located at the end of the laser ICF facility, which is one of the key components of the SG facility. In the process of laser target shooting, a variety of large-caliber optical elements are contained in the final optics assembly which has many essential functions, such as frequency conversion, harmonic separation, focusing transmission, beam sampling and so on. The effect of high-power laser will lead to the damage of the optical components, and optical components need to be frequently replaced and maintained. Linear replacement units (LRUs) of SG facility are formed by these optical components and their replacements [2].

The surface cleanliness of optical components is significant conditions for high throughput of the laser in high-power laser devices [3, 4]. The surface contaminants of optical components can affect the beam quality and reduce the damage resistance of optical components seriously, which result in decreasing load capacity of the laser driver [5, 6]. Researchers suggested that suspended particulates were a pivotal cause in the surface contamination of optical components [7, 8]. Optical components need to avoid that the optical components directly expose in the air containing dust particles when they are removed and installed. It is necessary to grasp closed loop cleaning control of the online clean maintenance process. The crystal loop case for optical crystal module in final optics assembly can not only realize the rapid assembly of the optical components, but also can avoid the pollution of suspended particles during the online maintenance process of the optical components [9]. Pryatel et al. [10] analyzed the cleaning process, treatment methods and cleaning equipment in NIF and provided the theoretical basis for cleaning transportation equipment design. Wang et al. [11] investigated the lower part of the LRUs module in the laser device for the position monitoring technology and provided the conditions for the module’s installation process. However, these researchers did not directly involve the problems about calculating flow field and designing the cleaning module. Wong [12] analyzed the production mechanism of particle contamination in high-power laser system and provided a theoretical basis for the engineering design of the device. The crystal loop case had been designed for the optical components’ assembly in NIF. The loop case design for the optical components’ assembly met the requirement of online clean maintenance for large aperture optical component modules [13]. In order to ensure the well-ventilation in crystal loop case and realize effective online removing of optical elements, it is necessary to analyze the internal flow field state when the loop case is designed. To confirm that the change regulations of the cleaning state inside the loop case can be mastered, the design of the loop case need to be optimized [14]. So far, in the development of high-power laser device, the design and analysis of the loop case for optical crystal module in final optics assembly were seldom reported. The clean state analysis of the loop case and its verification tests are the necessary prerequisite for cleaning design of the high-power laser device and its maintenance equipment. It is of great significance to realize fully the closed loop clean control of systems and promote the development of ultra-clean manufacturing in high-power laser device.

2 The Design of Loop Case and Analysis of Air Flow Field

2.1 The Design of Loop Case

On the one hand, in order to achieve a clean control of the ShenGuang-III facility during online maintenance, the clean gas is passed into the throttle plate of the FOA. Meanwhile, to isolate external gas pollution, a positive pressure environment can be formed by the clean gas in the intersection area between the FOA and loop case. On the other hand, a stable and uniform flow field in the loop case can be formed by the case design. Flow field can efficiently remove the particle contamination from the loop case production during operation and prevent the optical components from secondary contamination [15]. The shape of the loop case is designed to be cuboid due to the structural characteristics of the final optics assembly interface and the site maintenance conditions. The cleanliness level in high-power laser facilities environment is about 300 thousand level. The maximum wind speed is less than 1 m/s in FOA except the vents position, and the average wind speed in indoor is of the range from 0.2 to 0.5 m/s. The 3D model of the loop case for optical crystal module in FOA is shown in Fig. 1. In order to reduce the reflux near the outlet position, more conducive to the rapid discharge of pollutants here reduces the possibility of secondary contamination of crystal elements by suspended particulate pollutants in the loop case. Two vent ports at the upper and lower ends of the loop case are designed except the operation port. Therefore, there are three ports, namely the upper, middle and lower ports in the end of the designed loop case.
Fig. 1

3D model of loop case

2.2 Simulation Model and Critical Gas Flow Velocity

In this paper, the study performed the diffusion effect simulation of external gas inside the loop case. Besides, it analyzed the isolation effect of positive pressure clean gas resisting the outside polluted gas too. In the simulation, the method of expanding region was used to deal with the environmental condition, which meant extending the air flow field at the loop case opening. The simulation model after the grid partition is shown in Fig. 2. According to the actual working conditions, the loop case connects the optical frequency-doubling crystal module of FOA during maintenance. In these conditions, the air entering the loop case through the frequency components transforms cleaning air after purifying and drying. The gas flow is regarded as an incompressible flow in the simulation. This paper focus on the distribution of the flow field after gas flow stabilization in the study of fluid motion model; therefore, the steady-state turbulence method is chosen as the analysis model [16]. The nonslip wall function is used to deal with the near-wall surface, the coupling method of the velocity and pressure is chosen by SIMPLE algorithm, and the second-order upwind difference is adopted in the simulation. Moreover, the convergence speed is adjusted by using the under-relaxation factor.
Fig. 2

Simulation model after meshing

The above model is simulated calculation by changing the inlet speed of the airflow. The streamline diagram shows clean gas in a steady state which is set as 0.2 m/s at the inlet speed in Fig. 3. According to the inlet area of cleaning gas, the inlet velocity of 0.2 m/s is converted into the intake rate of flow of 650 L/min. Figure 3 shows that the clean gas flows into the component evenly from the throttle plate. The eddy current motion is formed inside the components when the gas flow state is stable. Then clean gas enters into the loop case through the crystal mounting hole, and the laminar flow movement is formed in the loop case where there is not obvious eddy current phenomenon. Finally, the cleaning gas is discharged from the opening loop case outside the case.
Fig. 3

Stream line of clean air

So as to evaluate the isolation effect of the positive pressure clean gas on the outside pollution air, the study used the measuring method of the normal velocity on the cross section in this paper. The definition of the outward direction of the clean transport box is the negative direction of the x-axis when the simulation model is established. In the simulation process, if the maximum normal velocity is less than zero, all gas on the cross section moves outside the crystal loop case at this time. In other words, the outside pollution gas cannot enter the loop case from this cross section. The gas flow field at different inlet velocities is simulated. Three exhaust holes are set up to three monitoring surfaces at the opening of the loop case. The maximum velocity in the X-direction on the three monitoring surfaces is defined as outlet 1, outlet 2 and outlet 3, respectively. It performed the simulation of the gas flow field at different inlet velocities. The X-direction velocity curve of each monitoring surface under different inlet speeds is shown in Fig. 4.
Fig. 4

Max. velocity on the monitoring plane

Figure 4 shows that the maximum velocity on the monitoring plane correlates well with the inlet speed. Due to the impact of the loop case structure, the slight fluctuation of the maximum velocity on the monitoring plane ranges from 0.25 to 0.4 m/s. When the inlet speed is less than 0.2 m/s, the maximum velocity of outlet 2 is greater than zero. Therefore, the outside polluted gas can enter the loop case from outlet 2. When the inlet speed exceeds 0.2 m/s, the maximum velocity of outlet 2 is less than zero and the outside gas cannot enter the loop case. Meanwhile, by analyzing average velocity of each monitoring surface, the results show that the average velocity of each monitoring plane is linearly related to the inlet velocity of the cleaning gas. The analysis indicated that average velocity of each monitoring plane was linearly related to the inlet velocity of the cleaning gas. As the inlet velocity increases, the absolute value of the average velocity increases. The maximum velocities of the three monitoring planes are all less than zero when the inlet speed is bigger than 0.2 m/s. The result indicated that clean gas could isolate outside pollution. Therefore, the speed of 0.2 m/s is the critical air velocity to guarantee the cleaning state of the loop case in maintenance.

3 The Effect of Air Intake on Cleanliness

With the variation of the air inflow of positive pressure clean gas in final optics assembly, the trajectories of particles can be changed significant correspondingly. The intake air volume can affect the volume of the eddy current area inside the loop case and amount of the eddy current. The eddy current resulted in irregular diffusion movement of suspended particulates [17]. If the suspended particulates cannot be removed from the inside of the loop case, they will remain on the surface of the crystal elements. The suspended particulates would induce and aggravate the damage of the optical components under strong laser light [18]. The damage of optical elements is also a significant source of particle contamination. From the above analysis, the result can be found that optical elements can effectively isolate the outside polluted gas when the inlet speed reaches 0.2 m/s or intake capacity is 650 L/min. In addition, in order to remove surface debris at a rate of nearly 100% efficiency (> 60 μm), some scholars had studied a 1-s high-speed (76 m/s) air pulse from commercially available “air knife” crossing the mirror to its surface [19]. Nevertheless, in the actual work process, the particles outside the loop case will also pollute the crystal components. The particles from residual materials and production in the components during the operation are also likely to diffuse into the loop case. The particles remain inside the loop case under the action of positive pressure cleaning gas and further result in contamination of the crystal elements surface. Therefore, it is necessary to further analyze the particles changes inside the loop case with the intake air of the clean gas and determine the optimum air intake for the minimum residual particles in the loop case under the condition of isolating positive pressure protection.

To confirm improvement of the calculation efficiency, the connection surface of the loop case and the final optics assembly are set as the particles initial position. The cleanliness of the main equipment is required to be controlled to level 10, and the diameter of the particulate matter contained in the laser channel is less than 5 μm. Therefore, only the suspended particulate particles with the diameter of 0.1–5 μm are considered in this paper. Firstly, the initial state of particle phase is defined by compiling files with MATLAB software. The particles trajectory is tracked and calculated under different inlet velocities by inputting 231 particle sources with a diameter of 1 μm. And these particles are uniformly lattice distributed along the interface of the crystal loop case. Then, the trajectories of particles under the action of gas in the loop case in diverse initial positions are tracked. Finally, the relationship between the overall situation of particles and the inlet velocity of the clean gas is determined. In Fig. 5, space suspension refers to the number of suspended particles remained in the loop case. The surface capture refers to the number of particles deposited on the surface of the crystal. The total residue refers to the total number of particles remained in the loop case, which means the sum of suspended particles and deposited particles in space.
Fig. 5

Relation between the inlet velocity and the amount of residual particle

Figure 5 shows that the number of particles trapped on the surface is zero and the number of suspended particles is less when the inlet velocity is in the range of 0.2–0.25 m/s or 0.45–0.5 m/s. Once the inlet velocity exceeded 0.5 m/s, the number of particles trapped on the surface increases dramatically. The inlet velocity increase will increase the probability of the particles depositing on the surface of the crystal elements. Therefore, taking the external gas isolation effect into account, the intake velocity of 0.45–0.5 m/s or intake air volume of 1464–1626 L/min is chosen as the optimum parameter. Because the inlet velocity is too small to effectively remove the particulate matter in the loop case and also cannot form the isolation of the external pollution gas. When the inlet velocity is too high, the swirl area and the number of swirls in the loop case will increase, which aggravates the diffusion cycles of particulate pollutants, and it is not conducive to its cleanliness. So the wind speed is 0.45–0.5 m/s when the particles contamination in the loop case is the least.

4 The Influence of Crystal Mounting Sequence on the Clean State of the Loop Case

Crystal mounting sequence needs to be considered for an excellent state of the cleanliness in the crystal modules of the FOA. The micro-friction between crystal module and the mechanical component can result in the particles generation. Large particles can deposit at the module bottom under the action of their own gravity, and small particles may be blown into the loop case through junction surface in the role of gas flow field. Those small particles can deposit on the surface of the optical crystal under the action of non-uniform flow and form a new source of the contamination. In order to analyze the effect of the crystal mounting sequence on the clean state of the loop case, the crystal elements in the loop case are numbered, respectively. Crystal 1 is away from the target sphere. Crystal 2 is in the middle of the loop case, and crystal 3 is close to the target sphere, as shown in Fig. 6. Inlet velocity is 0.5 m/s, and the diameter of solid particles is 1 μm; the trajectory of particles inside the loop case is simulated with various simulation models in different installation sequences. As the position of three optical crystals is different, the fluid domain is modeled when an optical crystal is firstly installed. As listed in Table 1, three simulation models were presented according to the first crystal installed due to different location of the three optical crystals. Model 1 means that only crystal 1 is installed into the assembly. Model 2 means that only crystal 2 is installed into the assembly. Model 3 means that only crystal 3 is installed into the assembly.
Fig. 6

Number of crystal module

Table 1

Simulation models for various sequences of installation

Model

Description

1

Only install crystal 1 into the assembly

2

Only install crystal 2 into the assembly

3

Only install crystal 3 into the assembly

4

Install both crystals 1 and 3 into the assembly

5

Install both crystals 1 and 2 into the assembly

The grid division of three simulation models and simulation results of the models are shown in Fig. 7; the result can be found that the trajectories of the particles differ from different installation sequences. Meanwhile, the amount of particles trapped on the surface differs from suspended particles. Through the comparison and analysis of the three models, the result can be noticed that the amount of particles trapped on the surface is zero and the amount of suspended particles is the least in model 1. The trajectory of particles in the model 1 also has excellent parallelism. Therefore, in order to maintain the best cleaning state in the loop case, crystal 1 should be installed first when the optical crystal is maintained online.
Fig. 7

Motion trails of particles in various models. a Model 1, capture: 0, suspension: 4. b Model 2, capture: 1, suspension: 11. c Model 3, capture: 3, suspension: 13

Then, it is necessary to further analyze the influence of the mounting sequence of the crystal 2 and the crystal 3, which are on the trajectories of the particles after determining an installation scheme for the model 1 of the crystal 1. Model 4 means that crystals 1 and 3 are installed inside components sequentially. However, the model 5 means that crystals 1 and 2 are installed inside components sequentially. Models 4 and 5 are simulated, respectively, and the simulation results are shown in Fig. 8. The result can be found that the number of particles deposited on the surface of model 4 is less than model 5. The trajectory of the particles is parallel to the x-axis. There are a large number of vortex areas and a significant eddy current at the inlet of the loop case in model 5. There is no obvious eddy current at the inlet of the loop case in model 4 except the eddy current region caused by a square column flow. The particles will spread in all directions in the loop case under the action of the eddy current. Therefore, the number of particles deposited on the crystal element surface is more than the number of model 5. Through the above analysis, when the optical module is in the online clean maintenance, the sequence can be found that crystal 1 should be installed into the final optics components first and then crystal 3 should be installed. Finally, crystal 2 is installed into the final optics assembly.
Fig. 8

Motion trails of particles in the loop case. a Model 4, capture: 1, suspension: 5. b Model 5, capture: 9, suspension: 5

5 Experimental Validation of Clean State of the Loop Case

The detection of particle contamination is a direct way to determine the source and state of contamination [20]. In order to validate the simulation results, the clean state detection and verification tests of the loop case are carried out in a 100-level clean laboratory. As shown in Fig. 9, equipment such as the loop case, throttle, crystal components module and laser particle counter is used for building experimental system. In the experiment, the crystal element module is a frequency-doubling crystal module which is used in the target range. The clean gas is supplied by the special gas pipeline, and the airflow velocity at the opening of the loop case is detected by the anemometer.
Fig. 9

Experimental setup for the cleaning status monitor

The inlet speed of the clean gas is set as 0.2 m/s, and the heat-sensitive wind speed probe is placed at the opening of the loop case to measure the airflow speed when the measuring state is stable. Fifteen monitoring points are set at the opening places, and the average value of each monitoring point is measured on multiple times. Finally, the velocity averages of all the monitoring points are taken as the speed at the of the loop case opening. Figure 10 shows the relationship between the average velocity of the exit and the inlet velocities from the experiment. The experimental results show that with the increase in inlet velocity, the absolute value of the average velocity at the outlet increases accordingly, which has the same conclusion that the average velocity at each monitoring surface is linearly related to the inlet velocity of clean gas. Meanwhile, it can be seen that with the increase in inlet velocity, the positive pressure clean gas for the crystal loop case opening of the external air flow protection and isolation effect gradually becomes better, which is consistent with the simulation results of the positive pressure clean gas.
Fig. 10

Relationship between the average velocity of the exit and the inlet velocities

According to the simulation analysis of particles in the loop case, it can be concluded that the cleaning state of the loop case is the best state when the air intake volume is between 1464 and 1626 L/min. To verify the simulation results, the experiments on the number change of particles under different inlet speeds were carried out. Two laser particle counters were used in the experiment. One was used to measure the number of particles entering the loop case named Nin, and the other was used to measure the number of particles exhausted from the loop case named Nout. The ratio of Nout to Nin is regarded as the judging standard of the cleaning state inside the loop case. The crystal modules were installed into the loop case, and two laser particle counters sampling heads were placed at the import and export in the loop case. Nin and Nout are measured under different inlet speeds by changing the intake of clean gas. Six groups of experiments were performed at each inlet speed group under the condition that the entrance speed was kept constant. The result reaches the average of experimental data. Figure 11 shows the relationship curve about the ratio of Nout/Nin and inlet velocity. When the inlet velocity is 0.45 m/s, the ratio of Nout/Nin reaches the maximum value. The experimental results have the same conclusion with the optimal inlet velocity selected in the simulation analysis, which verifies the correctness of the simulation results of gas–solid two-phase flow. Both too high and too low inlet velocities will lead to an increase in the number of eddies. Under the action of eddy current, the suspended particulate matter will spread irregularly in the crystal loop case and it is left inside the crystal loop case and destroys the clean state.
Fig. 11

Relationship between the ratio of I/O and the inlet velocities

6 Conclusions

In this paper, the influence of the intake air volume on the clean state inside the loop case is obtained by the simulation. The simulation results show that the minimum air intake volume is 650 L/min which can isolate the outside polluted gas. Based on the theory of gas–solid two-phase flow, the optimal air intake volume is chosen as range from 1464 to 1626 L/min which can keep excellent cleaning state inside the loop case by using the simulation results of the solid particles’ trajectory. It determines the optimum installation sequence of crystal elements in the loop case. Finally, the study sets up an experimental detection system, which detects the clean state of loop case. The simulation results were verified by experiments. The results of this study can provide a theoretical basis for the engineering design of the loop case and formulation of clean process.

Notes

Acknowledgements

This research work was jointly supported by the State Key Program of National Natural Science Foundation of China (Grant No. 51535003) and the National Natural Science Foundation of China (Grant No. 51575138).

References

  1. 1.
    Lindl J et al (2014) Review of the national ignition campaign 2009–2012. Phys Plasmas 21(2):020501CrossRefGoogle Scholar
  2. 2.
    Larson DW (2004) NIF laser line-replaceable units (LRUs). Proc SPIE 5341:127–136CrossRefGoogle Scholar
  3. 3.
    Bude J, Carr CW, Miller PE (2017) Particle damage sources for fused silica optics and their mitigation on high energy laser systems. Opt Express 25(10):11414CrossRefGoogle Scholar
  4. 4.
    Feigenbaum E, Raman RN, Cross D et al (2016) Laser-induced Hertzian fractures on the exit surface of silica glass deposited with metal micro-spheres. Proc SPIE 10014:1001406CrossRefGoogle Scholar
  5. 5.
    Kohli R, Mittal KL (2015) Developments in surface contamination and cleaning. Volume 8: cleaning techniques. Elsevier, Amsterdam. ISBN: 978-0-3232-9961-9Google Scholar
  6. 6.
    Sun X, Lei Z, Lu X et al (2014) Mechanism of original damage of thin optical components induced by surface particle contamination. Acta Phys Sin 63(13):134201Google Scholar
  7. 7.
    Palmier S, Tovena I, Courchinoux R et al (2005) Loser damage to optical components induced by surface chromium particles. Proc SPIE 5647:156–164CrossRefGoogle Scholar
  8. 8.
    Kohli R, Mittal KL (2017) Developments in surface contamination and cleaning. Volume 9: Methods for surface cleaning. Elsevier, AmsterdamGoogle Scholar
  9. 9.
    Wang H, Xiong Z, Yao C et al (2016) Assembly modeling and error analysis of large laser optics. Proc CIRP 44:412–417CrossRefGoogle Scholar
  10. 10.
    Pryatel JA, Gourdin WH, Frieders SC et al (2014) Cleaning practices and facilities for the national ignition facility (NIF). Proc SPIE 9237:92372HCrossRefGoogle Scholar
  11. 11.
    Wang J, Lu C, Li S et al (2016) Simulation based assembly and alignment process ability analysis for line replaceable units of the highpower solid state laser facility. Fusion Eng Des 112:7–13CrossRefGoogle Scholar
  12. 12.
    Wong J, Ferriera JL, Lindsey EF et al (2006) Morphology and microstructure in fused silica induced by high fluence ultraviolet 3× (355 nm) laser pulses. J Non-Cryst Solids 352:255–272CrossRefGoogle Scholar
  13. 13.
  14. 14.
    Bai Q, Guo Y, Chen J et al (2016) Research and development of ultra-clean manufacturing. J Mech Eng 52(19):145–153CrossRefGoogle Scholar
  15. 15.
    Moses EI (2010) The national ignition facility and the national ignition campaign. IEEE Trans Plasma Sci 38(4):684–689CrossRefGoogle Scholar
  16. 16.
    Shih TH, Liou WW, Shabbir A et al (1995) A new k–epsilon eddy viscosity model for high reynolds number turbulent flows. Comput Fluids 24(3):227–238CrossRefGoogle Scholar
  17. 17.
    Fan H, Liu S, He Z, Li X (2002) Large eddy simulation of flow field in vector flow clean-room. Tsinghua Sci Technol 7(3):326–330Google Scholar
  18. 18.
    Spaeth ML, Manes KR, Honig J (2015) Cleanliness for the NIF 1ω laser amplifiers. Fusion Sci Technol 69(1):250–264CrossRefGoogle Scholar
  19. 19.
    Gourdin WH, Dzenitis EG, Martin DA, et al (2005) In situ surface debris inspection and removal system for upward-facing transport mirrors of the National Ignition Facility. In: Proceedings of SPIE the international society for optical engineering, vol 5647, pp 107–119Google Scholar
  20. 20.
    Kohli R, Mittal KL (2010) Developments in surface contamination and cleaning. Volume 2: particle deposition, control and removal. Elsevier, AmsterdamGoogle Scholar

Copyright information

© International Society for Nanomanufacturing and Tianjin University and Springer Nature 2018

Authors and Affiliations

  1. 1.School of Mechanical and Electrical EngineeringHarbin Institute of TechnologyHarbinChina
  2. 2.Research Center of Laser FusionChina Academy of Engineering PhysicsMianyangChina

Personalised recommendations