Zhengxiang Shen

LAMP: a micro-satellite based soft x-ray polarimeter for astrophysics

The Lightweight Asymmetry and Magnetism Probe (LAMP) is a micro-satellite mission concept dedicated for astronomical X-ray polarimetry and is currently under early phase study. It consists of segmented paraboloidal multilayer mirrors with a collecting area of about 1300 cm2 to reflect and focus 250 eV X-rays, which will be detected by position sensitive detectors at the focal plane. The primary targets of LAMP include the thermal emission from the surface of pulsars and synchrotron emission produced by relativistic jets in blazars. With the expected sensitivity, it will allow us to detect polarization or place a tight upper limit for about 10 pulsars and 20 blazars. In addition to measuring magnetic structures in these objects, LAMP will also enable us to discover bare quark stars if they exist, whose thermal emission is expected to be zero polarized, while the thermal emission from neutron stars is believed to be highly polarized due to plasma polarization and the quantum electrodynamics (QED) effect. Here we present an overview of the mission concept, its science objectives and simulated observational results.

Using monodisperse SiO2 microspheres to study laser-induced damage of nodules in HfO2/SiO2 high reflectors

Nodules have been proved to play an important role in the activation of laser damage in 1.053 μm HfO2/SiO2 high reflectors. However, some damage test results revealed that the ejection fluences of some big nodules with height around 1 μm were abnormally high. To find the correlation between the surface dimensions of nodules and their susceptibility to nano-second pulsed laser radiation, monodisperse SiO2 microspheres with five different sizes were used to create engineered nodules in 1.053 μm HfO2/SiO2 high reflectors. The defect density of nodules that were created from SiO2 microspheres was purposely controlled to be around 20-40 mm2 and special care was taken to minimize clusters of SiO2 microspheres as less as possible. This enabled us to take a raster scan test and to get the statistical value of ejection fluences of these engineered nodules. The height and width dimensions of the engineered nodules, especially the discontinuity of nodular boundary, were measured by cross-sectioning of nodular defects using a focused ion-beam milling instrument. Based on the above information, the damage test results were interpreted from the aspects of electric field enhancement model and mechanical stability of nodular structures.

Using engineered defects to study laser-induced damage in optical thin films with nanosecond pulses

By creating nodules from artificial seeds, the damage behaviors of engineered nodules were systematically studied from experimental approaches. The seed diameters, seed absorption and film absorption were varied independently to uncover a single factors influence on the damage behavior of nodules. First, non-absorbing monodisperse SiO2 microspheres with five different sizes were used to create engineered nodules in 1.053 μm HfO2/SiO2 high reflectors that were prepared by EB process. Laser damage test results showed that the ejection fluences of nodules monotonically decreased with the increase of silica microsphere diameters. And to our surprise, nodules initiating from 0.3 and 0.6 μm silica seeds survived the maximum fluence of 170 J/cm2 (10 ns). Film absorption also has big influence on the damage behaviors of nodules. Compared to the nodules in low absorbing reflectors that were prepared by EB process, the nodules in high absorbing reflectors that were prepared by IAD process exhibited a much lower ejection fluences, although the seed diameters for the comprising nodules were same. Moreover, aluminum seeds were also used to create engineered nodules. Laser damage test results showed that the ejection fluences of nodules initiating from aluminum seeds were around 2 J/cm2 (10 ns), which is more than an order of magnitude less than the ejection fluences of nodules created from nonabsorbing silica seeds. This result revealed that the seed absorption played a very important role in the laser damage of nodules.

Laser damage resistance of dichroic mirrors at 532nm and 1064nm

HfO2/SiO2 dichroic mirrors were prepared using reactive electron beam evaporation process. The mirrors have high reflectance at 532 nm (S polarization) and high transmittance at 1064 nm (P polarization) for angle of incidence of 45°. Here we report the laser damage behaviors of the dichroic mirrors that were irradiated by 532 nm and 1064 nm nanosecond laser pulse separately. The influence of substrate polishing quality on the laser damage resistance of HfO2/SiO2 dichroic mirrors was also discussed. At 1064 nm, the nano-sized absorbers between coatings and substrate interface or subsurface are the precusors for creating damage. And the poor substrate polishing quality significantly decreases the laser induced damage threshold (LIDT). At 532 nm, two distinct damage behaviors initiating from visible nodules and nano-sized absorbers were observed. Nodules are ejected at low fluence but the ejected pits are very stable until quite high fluence around 20J/cm2 (1 ns). The nano-sized absorbers trigger damage at medium fluence around 14J/cm2 (1 ns), and this kind of damage grows very fast during subsequent irradiations. The substrate polishing quality has minor influence on the LIDT at 532 nm.

Study of laser induced damage of high reflector at 1064 nm

One TiO2/SiO2 high reflector with absorption of 300 ppm and two HfO2/SiO2 high reflectors with absorption of 40 and 4.5 ppm were fabricated using electron beam evaporation method. The influence of average absorption on laser induced damage threshold (LIDT) at 1064 nm of these coatings was studied using r-on-1 test mode. The LIDT of high absorbing TiO2/SiO2 coating is only 6.2 J/cm2 at 3 ns, whereas, two weak absorbing HfO2/SiO2 high reflectors got almost the same LIDT that was over 100 J/cm2 at 3 ns. Then a raster scan test method was employed to find the typical damage morphologies of these coatings with different absorption level. For TiO2/SiO2 high reflector with high average absorption, the representative damage morphologies are shallow pits that were probably caused by absorbing centers. However, for HfO2/SiO2 high reflectors with low average absorption, the dominant damage morphologies are micrometer-sized nodules ejected pits and the delamination initiating from the pits. The absorption of HfO2/SiO2 coatings is low enough to have minor influence on the laser induced damage.

LIDT of HfO2/SiO2 HR films by different test modes at 1064nm and 532nm

In order to evaluate the laser induced damage threshold (LIDT) of our HfO2/SiO2 high reflectance films prepared by reactive electron beam evaporation process at 1064nm and 532nm, the four popular test methods are performed according to the ISO 11254 and other relevant standards. The improvement of laser conditioning effect and influence of cumulative effect have been studied and estimated during the tests by comparing the deviations of the thresholds along the damage probability curves and relationship between the thresholds and the pulses number. Moreover, the details are carefully inspected during raster scan especially at 1064nm with all the >2μm nodules and damage sites marking and recording, then the ejection probability and growth rate of nodules are given. Note that attention should be paid to the submicron absorbing particulates at 532nm which are probable to trigger the damage.

Calculation and simulation of the uniformity of grinding removal in ring polishing

Ring polishing, also called continuous polishing, plays a very important role in the manufacturing of plane optical components with large aperture. This paper theoretically analyses the problem of calculation and simulation of the uniformity of grinding removal in ring polishing. By using the MATLAB software, a series of simulation figures are given. Firstly, the relative motion path on the polishing lap of a point on the workpiece is obtained by programming. From the simulation results it could draw a conclusion that the motion path is complicated when rotating speed ratio is not equal to one. Thus, its beneficial to the homogeneous material removal. Focus on the problem of the uniformity of grinding removal, then, this paper elaborates on the effect of material relative removal caused by factors such as rotating speed ratio and eccentricity theoretically. When the size of the workpiece is large enough, it will be outside of the optical polishing pitch lap, and this paper will also discuss the material removal on the whole surface in this case. All the calculation and simulation will guide practice process.

Fabrication of flat and supersmooth surfaces with bowl-feed polishing process

Low-scatter supersmooth optical components are wildly used for critical applications such as high power laser systems, mirrors for ring laser gyros, and low-scatter windows. In this paper, a bowl-feed polishing process with fine abrasives to get flat and extremely smooth surfaces is presented. We consider the machining of super-smooth surfaces as a chain consisted of some key nodes, not merely a polishing process. So after the samples of fused silica have been grinded, fine grinded and lapping previously, each step is controlled strictly, a bowl-feed polishing process is adopted to produce the supersmooth surfaces. During the whole polishing process, in order to achieve consistently reliable and meaningful values of both roughness and flatness, every effort is made to get the proper polishing parameters. Abrasives with different grain size distribution and the temperature are taken into account particularly. Two characters of the samples, roughness and flatness, are concerned about in this study. Process controlling by interferometer implies that surface shape is maintained to better than λ/15 RMS (λ=632.8nm), 100 mm in diameter for an extended periods of time, of the order of days. The surface roughness is tested by an interference microscopy. In best cases, the final test result of surface roughness is better than 0.8nm RMS.

Ray tracing method for the evaluation of grazing incidence x-ray telescopes described by spatially sampled surfaces

The nested grazing incidence telescope can achieve a large collecting area in x-ray astronomy, with a large number of closely packed, thin conical mirrors. Exploiting the surface metrological data, the ray tracing method used to reconstruct the shell surface topography and evaluate the imaging performance is a powerful tool to assist iterative improvement in the fabrication process. However, current two-dimensional (2D) ray tracing codes, especially when utilized with densely sampled surface shape data, may not provide sufficient accuracy of reconstruction and are computationally cumbersome. In particular, 2D ray tracing currently employed considers coplanar rays and thus simulates only these rays along the meridional plane. This captures axial figure errors but leaves other important errors, such as roundness errors, unaccounted for. We introduce a semianalytic, three-dimensional (3D) ray tracing approach for x-ray optics that overcomes these shortcomings. And the present method is both computationally fast and accurate. We first introduce the principles and the computational details of this 3D ray tracing method. Then the computer simulations of this approach compared to 2D ray tracing are demonstrated, using an ideal conic Wolter-I telescope for benchmarking. Finally, the present 3D ray tracing is used to evaluate the performance of a prototype x-ray telescope fabricated for the enhanced x-ray timing and polarization mission.

Improving the surface metrology accuracy of optical profilers by using multiple measurements

Abstract. The performance of high-resolution optical systems is affected by small angle scattering at the mid-spatial-frequency irregularities of the optical surface. Characterizing these irregularities is, therefore, important. However, surface measurements obtained with optical profilers are influenced by additive white noise, as indicated by the heavy-tail effect observable on their power spectral density (PSD). A multiple-measurement method is used to reduce the effects of white noise by averaging individual measurements. The intensity of white noise is determined using a model based on the theoretical PSD of fractal surface measurements with additive white noise. The intensity of white noise decreases as the number of times of multiple measurements increases. Using multiple measurements also increases the highest observed spatial frequency; this increase is derived and calculated. Additionally, the accuracy obtained using multiple measurements is carefully studied, with the analysis of both the residual reference error after calibration, and the random errors appearing in the range of measured spatial frequencies. The resulting insights on the effects of white noise in optical profiler measurements and the methods to mitigate them may prove invaluable to improve the quality of surface metrology with optical profilers.