Search

To achieve an ultrahigh-resolution for soft X-ray beamlines, the slope error of a highly precise grating is required on the level of 0.1 μrad root-mean-square (RMS) under thermal loading. To realize the goal, a specially designed 25-actuator optical surface bender for the gratings and mirrors is developed and operated at Taiwan Photon Source (TPS) [1]. In this paper, the construction and operation of the in situ LTP measuring system is described[2]. This LTP consists of a switchable optical reflection system that let the LTP can switch to measure horizontal or vertical mounting mirrors/gratings in the beamline. The other is a low optical distortion and bakeable to 120 ˚C glass viewport which is used for the ultra-high vacuum[3,4] interface for the beamlines optics and LTP. The surface slope error being reduced down to 0.1 and 0.15 μrad (RMS) by the 25-actuator bender without/with the glass viewport as verified by the in situ LTP measurements in the beamline.

The goal of an in situ Long Trace Profiler (LTP) is to adjust the mirror to 0.1 μrad Root Mean Square (RMS) under thermal load. Here we introduce the measurement configuration for in situ LTP. To avoid lens aberration, the moving optical head keeps the optical paths constant, and the reference beam is used to the correct of the unavoidable air bearing errors. The window glass in this test has a rather high optical quality, with a flatness of 1/150 (RMS) over 120 × 20 mm. The optical quality of the window was specified to be ± 1 μrad slope distortion in an aperture length of 100 mm. The window glass deformation for the air pressure was calculated by the Finite Element Method (FEM) software (ANSYS). The window glass deformation results can be fitting by the Zernike polynomial, and then bring it into the sequential optical ray tracing software (ZEMAX), and evaluating the window glass effect on the LTP measurement results. By this approach, we found that this has a constant error. Thus, the window glass air pressure error can be effectively removed from the measurement result to reveal the real mirror profile. Using the in situ LTP measuring result and the data iteration process, the bendable mirror can control the optical surface locate profile and thereby minimize the thermal distort effect. The slope error will be reduced to 0.1 μrad at the thermal load.

To achieve an ultrahigh resolution for soft X-ray beamlines at Taiwan Photon Source (TPS), the slope error of a highly precise grating is required on level of 0.1 μrad (rms) under thermal loading with various curvatures. On the beamline, some optics are usually operating under high power density from undulator magnet, the thermal load will introduce a thermal bump on the optics profile and degrade the beamline performance, such as energy resolution and beam size. To realize the high resolution goal, a specially designed bender with 25 actuators for the grating is designed and a In situ long trace profiler (LTP) with precision of 0.1 μrad (rms) has been developed to measure the mirror profile in soft X-ray beamlines. This article introduces the design and construction of in situ LTP. It can provide a feedback guideance for the adjustment of actuators of bender mechanism to achieve the optium profile. A suitable adjustment procedure from the input of in-situ LTP , performance of bender and energy spectrums are presented. There are several benders as the active mirrors and active gratings in operation in TPS 41A resonant inelastic X-ray scattering (RIXS) and TPS 45A angle-resolved photoemission spectroscopy (ARPES) beamlines. In the meantime, three in situ LTPs have been developed to monitor the grating profile under the thermal load in the beamlines. They are providing a feedback to measure the surface figure and to find the optimal surface profile. They would increase our efficiency to reach the energy resolving power of 35,000 and 28,000 in the RIXS and ARPES beamlines, respectively.

The project “High loading precision rotation stage design for synchrotron radiation mirror measurement” aims to provide an ultra-high-precision heavy-duty rotation stage and X-ray mirror interference optical measurements. Since the shape of an X-ray lens is very different from that of the general visible optical lens, the measurement system is very different from the general visible light optical measurement system. This paper describes a high-load precision rotating platform for obtaining stitching interferometer measurements for a synchrotron radiation mirror. The synchrotron radiation mirror is usually rectangular, and the length is greater than the interferometer measurement size. Therefore, for the mirror measurement, the stitching method is usually used to obtain synchrotron radiation mirror measurements. The interference measurements are obtained at different positions. In order to obtain the measurements, the center line of the interferometer must be perpendicular to the tangent of the mirror surface, so that appropriate interference fringes can be obtained. As the mirror radius becomes smaller, the interferometer rotation angle sensitivity increases. Development of the stitching interferometer high-load precision rotating platform design target requires an angle rotation resolution

A long trace profiler in NSRRC is used to develop a bendable mirror, for the mirror-mounting mechanism and to inspect the mirrors of TPS beamlines. We upgraded the air bearing, the motor and gear, the penta-mirror, the CCD and the software. ELCOMATT 3000 is our calibration reference. To maintain constant the measurement of the optical path, we adopted a scheme for a moving optical head that decreases the various optical paths through a focusing lens to avoid lens homogeneity. The measurement environment (temperature, vibration, air turbulence) is effectively controlled. In measuring a curved mirror (radius 9.7 m), the repeatability is below 0.1 μrad. This paper describes the upgraded performance and engineering details.

To achieve an ultrahigh resolution of a beamline for soft X-rays at the Taiwan Photon Source (TPS), the profile of a highly precise grating is required at various curvatures. The slope error could be decreased to 0.1 µrad (rms) at a thermal load with a specially designed bender having 25 actuators. In the meantime, a long-trace profiler (LTP) was developed in situ to monitor the grating profile under a thermal load; it consists of a moving optical head, an air-bearing slide, an adjustable stand, and a glass viewport on the vacuum chamber. In the design of this system, a test chamber with an interior mirror was designed to simulate the chamber in the beamline. To prevent an error induced from a commercial viewport, a precision glass viewport (150CF, flatness 1/150 λ rms at 632.8 nm) was designed. The error induced from the slope error of the glass surface and the vacuum deformation was also simulated. The performance of the optical head of the LTP in situ (ISLTP) has been tested in the metrology laboratory. The sources of error of this LTP including the linearity and the glass viewport were corrected after the measurement. For the beamline measurement, an optical head was mounted outside the vacuum chamber; the measuring beam passed through the glass viewport to measure the grating profile in vacuum. The measurement of the LTP after correction of the above errors yielded a precision about 0.2 µrad (rms). In a preliminary test, an ISLTP was used to measure the grating profile at soft X-ray beamline TPS45A. The measured profile was for the bending mechanism to optimize the slope profile. From the measured energy spectrum, the slope error of the grating was estimated with software for optical simulation to be about 0.3 µrad (rms), consistent with our estimate of the ISLTP. In the future, it will be used to monitor the thermal bump under a large thermal load. In addition, an ISLTP was used to monitor the properties of optical elements-the twist and radius in the beamline during the installation phase.

To ensure the high performance of beamlines in Taiwan Photon Source, the metrology of the beamline optical components is essential to realize the theoretical optical simulation. The design and performance of a long-trace profiler (LTP) with moving optical head is described. A LTP is used as a metrological tool for the assembly and correction of optical components at the beamline. For a water-cooled mirror, a LTP serves to optimize the surface profile to a slope error of 0.1 µrad (rms) and adequate thermal conductance. To correct a bendable cooling mirror in a twisted condition, the sagittal slope is decreased by a screw adjustment through the LTP measurement. For the double-crystal monochromator (DCM) with flow-induced vibrations, we proposed an approach to stiffen the bellows hose partially inside the vacuum chamber. From an investigation of the vibrational spectra of the DCM before and after the treatment with the aid of the autocollimator, the current approach is confirmed to be an effective solution for suppression of vibration.

The contact angles of aluminum melts at 973 K containing 0.78 and 3.8 wt% of copper on a sapphire surface are measured by extruding fresh droplets into high vacuum to avoid oxidation. The wetting is clearly affected by alloying, oxidation and time of contact. A discrepancy between the monolayer theory and present measurement is indicated.

The contact angle of pure aluminum melt at 973 K on sapphire (0001) surface is measured by forcing a fresh aluminum droplet out of an alumina syringe into a high vacuum of 10−4 Pa. A value of 90° is obtained if oxygen contamination at liquid-solid interface is avoided by limiting the exposure time of droplet to oxidative residual gases before contacting the substrate to

An active grating based on a novel optical concept with bendable polynomial surface profile to reduce the coma and defocus aberrations had been designed and proved by the prototype testing. Due to the low glass transition temperature of the glue and the difference of thermal expansion coefficient between the 17‐4 steel bender and silicon, the prototype distorted from flat polished condition when thermally de‐blocked the polishing pitch. To improve the thermal deformation of the active grating in the polishing process, a new invar bender and high curing temperature glue were adapted to glue a silicon substrate on the bender. After some tests and manufacturer polishing, it showed acceptable conditions. In this paper we will present the design and preliminary tests of the invar active grating. Meanwhile, the design and analysis of a new 17‐4 PH steel bender to be electro‐less nickel plating and mechanical ruling for a new beamline will also be discussed.

The Long Trace Profiler (LTP) is a precise angle measurement instrument, with a sensitivity and accuracy that can be in the sub-micron radian range. LTP characteristics depend on the particular LTP system schematic design, and the quality of components and assembly. The conditions of temperature, alignment, and mirror support during the measurement proccess vary between different laboratories, which influences significantly the test repeatability and accuracy. In this paper we introduce a direct comparison method to test the same object at the same point in the same environment at the same time by using two LTPs, which significantly increases the reliability of the comparison. A compact, portable LTP (PTLTP), which can be carried to different laboratories around the world, is used for comparison testing. Stability comparison experiments between the LTP II at the National Synchrotron Radiation Research Center (NSRRC), and the PTLTP of Brookhaven National Laboratory (BNL) reveal significant differences in performance between the instruments. The experiment is set up so that each optical head simultaneously records both its own sample probe beam and also the probe beam from the other optical head. The two probe beams are reflected from same point on the mirror. Tests show that the stability of the PTLTP with a monolithic beam splitter is 10 times better than the stability of the LTP II which has a separated beam splitter unit. A scheme for comparing scanning measurements of a mirror is introduced. Experimental results show a significant difference between the two LTPs due mainly to distortions in the optical components inside the optical head. A new scheme is proposed for further mirror comparison scanning tests.

Spherically bent silicon crystal x-ray analyzers have been employed in high-resolution inelastic x-ray scattering experiments to increase the counting efficiency due to the small cross-section of the inelastic scattering processes of interest. [1] In these bent analyzers, strain causes a distribution of lattice spacing, limiting the achievable energy resolution. Hence, the silicon wafers were diced using precision diamond saws into an array of ~1x1 mm2 blocks, and then acid etched to remove the saw damage, leaving blocks ~0.6x0.6 mm2 glued to a spherical concave substrate. With this method, meV energy resolution has been demonstrated with a bending radius of 6.5 m. [2] We seek to optimize the dicing process using the technique of deep reactive ion etching (DRIE) to develop highly efficient crystal analyzers. Ideally, each individual block subtends an angle that matches the acceptance (Darwin width) of the silicon reflection. This requires block sizes of about 500 μm2. DRIE offers the flexibility of selecting the block size, with finely controlled groove widths (i.e., minimal loss of material), and hence the possibility of controlling the energy width. We have made a prototype analyzer using DRIE with block size of 470 μm2, groove widths of 30 μm, and about 500 μm deep. The wafer was then bent and glued to a glass substrate with 2-meter radius. Tests showed encouraging results, with the DRIE analyzer performing at the 100 meV level. Details of the process and further refinements will be discussed.

A Long trace profiler II (LTP II) with a stationary optical head and a moving penta prism was set up in our new metrology laboratory. It is a clean room with plus or minus 0.1 degrees Celsius temperature control for the stable measurement of LTP II. Some tests on the stability and repeatability of the LTP II were performed. The long term drift is less than 1 (mu) rad in 15 hours. The repeatability in the forward and backward scan is about 0.2 (mu) rad. Mirrors with different kinds of bending mechanisms were evaluated by the LTP II, some comments on the mounting and bending of the mirrors are discussed.

This study presents a Kirkpatrick-Baez mirror system which includes two high-order polynomial bendable mirrors in a Taiwan Light Source (TLS) infrared beamline to create and adjust more accurately collimating images on a focusing point. The source of infrared rays is synchrotron radiation from a TLS bending magnet, so the surface of a vertical focusing mirror (VFM) is designed on an elliptical shape. The Runge-Kutta numerical method is used to compute the optimal high-order polynomial shape of the horizontal focusing mirror (HFM), to focus the horizontal arc source on the point image. The HFM and VFM using 17-4 PH stainless steel substrate without an electroless nickel plate are mechanically bent from planar to the desired fifth-order polynomial shapes with central radii of 3.74 m and 5.43 m by the application of equal couples, respectively. The mirror fabrication process, mechanical design, and the method of adjusting the mirror shape using the Long Trace Profiler measurement system are described. Finally, the roughness of mirrors is 3 angstrom RMS. After the mirrors have been bent, the slope error over 2/3 of clear aperture length (170 mm) was reduced to less than 6.3 (mu) rad RMS.© (1999) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

A new Kirkpatrick-Baez-type focusing mirror system for use in synchrotron radiation IR beamlines is designed and fabricated. This mirror system, which contains two fifth-order-polynomial-corrected cylindrical mirrors, can collect and focus the long arc shape IR source from the bending magnet into a nearly perfect point image. To fabricate these two uncommon mirrors, 17-4 PH type stainless steel substrates are chosen and mechanically bent from planar to the desired fifth-order-polynomial-corrected cylindrical shapes with central radii of 3.74 and 5.43 m. The root mean square (rms) roughness and the slope error of these two mirrors are measured to be 0.3 nm and less than 6.3 µrad, respectively. The method for calculating the polynomial coefficients of both mirrors as well as the mirror fabrication process, mechanical design, and the method for adjusting the mirror shape using a long trace profiler are presented.