Optics manufactured by mechanical grinding and polishing inevitably will bring surface/subsurface damages and defects during the machining process. Laser polishing has been demonstrated as a technique capable of achieving ultra-smooth surface with no damage and low-defects, but by far optics polished by this technology are only sufficient for illumination applications. To achieve high quality optics, high precision laser ablation has been proved to be a promising technology for shape correction. With pulsed CO2 laser, high precision laser ablation can be performed by direct evaporation of unwanted surface asperities. To acquire nanometer scale high precision ablation, an accurate control and meticulous adjustment of temperature should be needed. Herein, a mathematical model has been established to assist the understanding of the thermal mechanism of CO2 laser ablation and subsequently a series of simulations have been extended to investigate the phase change of evaporation. The temperature of fused silica irradiated by CO2 laser can be controlled via laser power and pulse duration. To achieve nanometer ablation depth, a gentle evaporation regime at low laser intensity is necessary. The results indicated that the ablation depth linearly depend on laser fluence and depth control levels of nanometer are obtainable with the control of laser fluence.
An ideal pitch button blocking process determines the level of workpiece deformation, especially the high-aspect-ratio optics, during the blocking process and process of polishing later. We have studied the pitch button blocking process by Finite Element Analysis (FEA) according to the thermoelastic equation. Meanwhile, the optimized pitch button blocking has been gotten by FEA which includes the thickness and material of blocking plate, as well as the radius, arrangement, elastic modulus and coefficient of thermal expansion of pitch buttons. The numerical simulation of Nd:glass (Ø100 mm×2 mm thickness) shows that the surface figure change (ΔPV) which is induced by the thermal stress during pitch button blocking process is influenced seriously by the thickness of blocking plate.
Different size polishing powder and different pH value ceria slurries were used to polish fused silica glass､K9 glass and Nd-doped glass on pitch plate. Material removal rates (MRR) of glass polished with different size powder and various pH value slurries, and textures of each sample were characterized. The results show that powder size has an effect on glass polishing performance: scratch densities increase with the increase of polishing powder size; surface textures become rougher with the increase of the size of polishing powder. The slurry pH value also affects glass polishing performance: MRR of fused silica glass are lowest under any pH value slurry while Nd-doped glass has the largest MRR; removal rates of all three kinds of glass will rise under both acidic and alkaline condition. Near neutral polishing environment and smaller size powder are useful for the surface polishing process. The results further reveal polishing mechanism and provide the guidance for glass surface process.
Dynamic Mechanical Analysis (DMA) and Bending Beam Rheometer (BBR) are applied to detect the property of the polishing pitch and get the creep compliance combining the craftwork of polishing. The simulation of workpieces’ surface in polishing is completed based on the Preston formula, Boltzmann superposition principle and Boltzmann superposition principle. A V-curve is gotten on studying the PV’s time history of the polished workpiece. Meanwhile, a confirmatory experiment whose parameter is agree with the simulation is completed, which the optimum result of PV, 0.3λ (λ=623.8nm) is similar with the optimum result of PV by simulation, 0.27 λ. A resemble simulated surface is gotten on the same PV with the experiment.
The effect of deep HF etching on the surface quality and figure of fused silica optics has been investigated systematically. Fused silica samples (100 mm in diameter x 10 mm thick) were manufactured using the conventional grinding and polishing process. These processed samples are etched with different removal depth. Initially, the surface quality of fused silica samples is characterized in terms of surface roughness and surface defects. Many digs not more than 1μm deep are emerged which originates from the micron grinding cracks and crack pits. These digs worsened the surface roughness and frosted the sample. While submillimeter subsurface damage exposed through etching appear as sparkling dots under the high power lamp. The average total length of millimeter scratches on single surfaces is over 200 mm. Not all millimeter scratches could be exposed until removal depth of up to 2 μm. Finally, the surface figure behavior during deep etching has also been figured out. Etching on the edge of the upper surface of samples placed horizontally went faster than on the inside parts. The surface of samples placed vertically assumed a more complicated removal distribution, which can be both explained in terms of "fringe tip effect". For the change of surface figure PV, the initial surface figure feature plays an important role as well as the etching removal distribution.
During continuous polishing, temperature is a significant source of processing uncertainty. Three work pieces of different kind material (K9, Nd:glass and ULE) were polished on 2.4m continuous polisher. It turns out that temperature difference has different influence on different material work pieces. It also indicates that temperature difference aggravates the processing uncertainy. The deformation caused by temperature difference is simulated using ANSYS. It shows that, with top-bottom temperature difference of 0.1°C, the deformation of Nd:glass, K9 and ULE are 0.444E-6 m (about 0.7025λ), 0.249E-6 m (about 0.3925λ ), and 0.105E-8 m (about 0.00166λ), respectively. With radial temperature difference of 0.1°C, the deformation of Nd:glass, K9 and ULE are 0.831E-7 m (about 0.1313λ), 0.465E-7 m (about 0.07348λ) and 0.196E-9 m (about 3.0973E-4λ), respectively. To explore the top-bottom temperature difference and radial temperature difference along the polishing surface, a small aperture Nd:glass and a large aperture Nd:glass in polishing have been measured using thermal infrared imager. The results showed that for Ø 260 mm × 26 mm Nd: glass, the radial temperature difference is about 0.1°C, while the top-bottom temperature difference is about 0.1°C ~ 0.21°C. Contrastively, for 810 mm×460 mm×40 mm Nd:glass, the radial temperature difference have reached 0.4°C, while top - bottom temperature difference ranges between 0.1°C ~ 0.27°C. When element gets larger, it will suffer greater temperature difference. These temperature differences are great enough to cause deformation far beyond the polishing accuracy required. Finally, methods are proposed to diminish the effect of temperature difference.
Smoothing is a convenient and efficient way to correct mid-spatial-frequency errors. Quantifying the smoothing effect allows improvements in efficiency for finishing precision optics. A series experiments in spin motion are performed to study the smoothing effects about correcting mid-spatial-frequency errors. Some of them use a same pitch tool at different spinning speed, and others at a same spinning speed with different tools. Introduced and improved Shu's model to describe and compare the smoothing efficiency with different spinning speed and different tools. From the experimental results, the mid-spatial-frequency errors on the initial surface were nearly smoothed out after the process in spin motion and the number of smoothing times can be estimated by the model before the process. Meanwhile this method was also applied to smooth the aspherical component, which has an obvious mid-spatial-frequency error after Magnetorheological Finishing processing. As a result, a high precision aspheric optical component was obtained with PV=0.1λ and RMS=0.01λ.
The figures of the polishing pad are of great significance for the figures of optical workpieces in the continuous polishing process. Three main factors which affect the figures of the pad, including the polishing pad creep deformation, the calibration plate grinding and the ambient change in temperature, are analyzed in this work. Processing parameters including the eccentricity between the polishing pad and the calibration plate, the calibration plate thickness, angular velocities of the pad are introduced in this analytic process. With this method, the figures of the polishing pad in the continuous polishing can be obtained. This work provides theoretical guidance for the deterministic processing of the continuous polishing process.
Continuous ring polishing is the key process in large aperture optical elements. The surface figure of polishing pad is inferred by the offline testing surface figure of workpiece. The defects, low processing efficiency and uncertainty processing time in traditional continuous polishing, the real-time monitoring method of polishing is proposed. The realtime monitoring system is set up based on the computer, the dynamic interferometer, a beam expanding system and a beam reflecting system. There are a workpiece and a glass monitoring plate placing in same ring. The surface figure of workpiece, monitored by the monitoring plate, synchronize with the surface of glass monitoring plate in Peak-Valley (PV) and POWER. The new method with simple structure is fast measuring and judgmental directly to the changes of surface figures. The results of real-time monitoring and surface figure converging on the workpiece are valid for continuous polishing through experimental validation.
The low surface laser damage threshold of fused silica components in high power laser systems such as NIF restricts the improvement of the output fluence of those systems. Once damage is initiated and grows under subsequent laser shots, the components will go unusable. Subsurface damage (SSD) introduced during manufacturing has been identified as a main damage initiator. A good knowledge of SSD and how manufacturing influences it is essential to optimize manufacturing processes for damage free optics. Using the magneto-rheological finishing (MRF) wedge technique of better accuracy attributed to a tip, we have characterized the subsurface damage on fused silica optical surfaces ground with loose Al2O3 abrasives of different sizes. Larger abrasives generates longer cracks and the number density of cracks decreases sharply with the depth for each size. Rogue particles account for the occurrence of trailing indent scratches. Addition of rogue abrasives into relatively small base abrasive extends SSD more deeply than that induced by rogue abrasives alone. The linear model, with the proportional coefficient 3.511, fits the relationship between SSD depth and surface roughness (SR) better than the quadratic polynomial one. We believe SSD depth relates to SR more statistically than following some specified physical law. The linear relationship between SSD depth and the abrasive size was also established. The abrasive size turned out not to be as a good indictor of SSD depth as SR.
In the inertial confinement fusion (ICF) system, the mid-frequency errors of optical components will cause high-frequency modulation and nonlinear gain of laser beams. In this paper, theoretical simulations and experiments have been designed and operated, aiming at studying the effects of slotting methods on mid-frequency errors in Nd-doped glass continuous polishing. Based on Preston formula, theoretical simulations focus on the effects of slotting methods on the mid-frequency errors. The simulation results show that different slotting methods will cause different mid-frequency errors, and square and logarithmic shape grooves are easier to obtain smaller mid-frequency errors. On the basis of simulation results, two groups of experiments are carried out to do the Nd-doped glass continuous polishing. The results show that the mid-frequency error gradually decreases with the decrease of the spacing between grooves. The results also show that square shape groove is easier to get a smaller mid-frequency error than circular shape groove, which verifies the theoretical simulation results.
Effect of the polishing plane vibration on large-size optical workpieces in continuous polishing is studied. The vibration equation was deduced based on the existence of inclination between the polishing plane and z-axis direction. Influences of different parameters, such as the inclination, rotation speeds of the polishing plane and workpiece, the eccentricity and workpiece radius, on the polishing plane vibration were simulated. The simulations results show that rotation speeds of the polishing plane and workpiece is the most significant factor. The chaotic vibration of the polishing plane increases with increasing rotation speeds differences between the polishing plane and workpiece. When differences are small, periodic ups and downs of the polishing plane occur with the increase of polishing time. Experiments verified the influence of rotation speeds differences on the polishing plane vibration. The vibration affects PV of large-size optical workpieces in continuous polishing.
The lifetime of optics in high power laser system is typically limited by both laser-initiated damage and the subsequent
growth of laser-initiated damage sites. The single- and multiple-shot irradiations for HfO2 /SiO2 high-reflective coatings,
deposited from hafnia and silica at 1064nm in nanosecond were investigated. It was found that when shot number
increased, the possibility of damage growth increased as well. The relationship between field distribution and damage
morphologies and inner structures was discussed to reveal mechanism of damage initiation. Additionally, the damage
morphologies under different laser fluence and shot numbers were characterized to discuss the damage growth
mechanism upon subsequent pulses. The tested results illustrated the absorbers which induced damages were random
distribution, and the second highest peak of field intensity at the fourth interface was high enough to induce the
micron-sized damage pits. It was found that defect density had a significant impact on the damage site whether growing
or not upon subsequent laser pulses. Additionally, the growth resulted in delamination, and in turn delamination
accelerated damage growth, finally the catastrophic damage happened.
Experimental and theoretical progress on multiple wavelength laser induced damage of multilayer beam
splitters is reviewed. Test method for multiple lasers-induced damage thresholds were proposed based on
ISO 11254. Single and multiple laser-induced damage performance of the splitters, including damage
probabilities, damage thresholds and damage morphologies, were investigated respectively in order to
obtain better understanding of the damage mechanism at 355, 532 and 1064 nm. A judgment criterion for
coupling effect of different lasers in inducing damages was proposed and the detailed coupling efficiency
was also obtained. The small absorbing particle model and defect statistical model were put forward to
explain the experimental results. In addition, damage performance when the lasers arrived at the films with
nanosecond-delay in time was researched. Generally, the third harmonics play a key role in the damage
Coupling effect between 355 nm laser and 1064 nm laser in damage initiation and morphology formation are investigated on two different coatings prepared with Hf/SiO2 and HfO2/SiO2 respectively. It was found that materials had little influence on the couple effect. When extra 1064 nm pulse energy is low, 355 nm laser induced damage thresholds of both coatings increase because of laser conditioning and then when 1064 nm pulse energy is high enough, 355 nm LIDTs decreased. Damage morphologies are also studied to explore the damage mechanism at respective wavelength. For the entirely different electric field distribution, 355 nm laser induced damages are mainly from nanometer-sized absorbers at upper interfaces while initiators for 1064 nm laser locate at substrate-coating interface or substrate subsurface. Under simultaneous illumination, the sensitive defects are still the precursors, but the damages are more catastrophic compared with damages induced by 355 nm laser only and they also show representative damage characteristics induced by single laser, namely 355 nm laser induced small pits and 1064 nm laser induced large delamination. Further studies also show that delamination area grows with the increase of pit numbers induced by 355 nm laser at fixed 1064 nm laser fluence. A possible mechanism was proposed to interpret observed delamination area growth phenomenon.
The multilayer dielectrics (MLDs) for broad bandwidth 800nm pulse compression gratings were
fabricated with optimized design by electron beam evaporation using three different kinds of materials
(Ta2O5/SiO2/HfO2), which had more than 99% reflectance with bandwidth larger than 160nm around
the center wavelength of 800 nm and high transmission at the exposure wavelength of 413nm.
Laser-induced damage behaviors of the mirrors were investigated. It was found that the laser-induced
damage threshold (LIDT) of the samples could reach 1.0J/cm2 and 2.0J/cm2 in the normal beam (57
degrees, TE mode) at pulse duration of 50fs and 120fs, respectively. The depth information of the
damage sites at these two cases was explored by atomic force microscope (AFM). The reason of the
sample having so high LIDT was also discussed in this paper. The MLDs provide a solid base for the
high laser threshold 800nm pulse compression gratings and may open a new way for broad bandwidth
800nm reflectance coatings used in the ultrashort pulse laser system.
According to the parameter requirements of a graded reflectivity mirror with a Gaussian profile, the layer structure and the mask pattern are designed using a graded-thickness middle layer. The mask and the automatic mask-switchover equipment are designed considering the actual requirement of the thin films and the specific deposit facility. The uniformity of the layer thickness is analyzed. The measurement results indicate that samples prepared with this technique are basically in accordance with the design parameter. The scattering effect between the material molecules and the mask, thickness errors, and the alignment error between the mask and the substrate are the main factors that influence the deposit result.
To understand mechanisms underlying laser-induced damage of BK7 and fused silica, we calculate the temperature field of the substrates with CO2 laser irradiating at a given laser power and beam radius. We find that the two glasses show different thermal behaviors. A model is developed for estimating the time t to heat the surface of the substrates up to a particular temperature T with cw CO2 laser irradiation. We calculate theoretically the duration t that the samples are irradiated, from the beginning to visual catastrophic damage, with the assumption of damage threshold determined by the critical temperature. The duration t that the samples are irradiated, from the beginning to visual catastrophic damage, is investigated experimentally as well. Here we take the melting point or softening point as the critical temperature, given the thermomechanical coupling properties, which is enough to cause damage for BK7. Damage features are characterized by the sound of visual cracks. Finally, we calculate stresses induced by laser heating. The analysis of stress indicates that the damage of BK7 is due to the stresses induced by laser heating.
We calculate the transient and steady state temperature field of CO2 laser conditioning of fused silica using thermal transporting model. We find that surface temperature at the beam center grows rapidly at the initial heating stage both for the linear thermal transport and nonlinear heat transport, then grows very slowly approaching the stationary value. The axial thermal gradients of nonlinear thermal transport, accounting for the increased thermal conductivity at high temperature, are more smooth than those of the linear case.