The first large Fabry-Perot etalon (Ø35 cm) of the VTF instrument was coated successfully using IBS technique. The High Reflective (HR) coatings need to meet the reflectivity specifications (95 +/- 1%) over the entire wavelength range 520-870 nm and the entire aperture (Ø25 cm) and also preserve the plate's flatness and airgap uniformity between the two platesto be better than 3 nm RMS. The change of the figure error of the individual faces after HR coating was exceptionally small: For plate 1 (upper) it changed from 1.7nm RMS before coating to 2.12 nm after coating, no change at all for plate 2 (lower).
Due to the increase of astronomical projects and of their instruments, the request of large optics with higher optical performances does not stop growing. An important step in the manufacture of these optics is the deposition of high-precision optical coatings.
To answer to this request we developed coatings working at different angles of incidence and spectral ranges on large surface:
- anti-reflective coatings for large lenses with strong curvatures,
- dichroic coatings with sharp transition for large optics.
Main results will be presented on the basis of several examples of realization.
The Visible Tunable Filter (VTF) is a diffraction-limited narrowband tunable instrument for imaging spectropolarimetry in the wavelength range between 520 and 860 nm. It is based on large-format Fabry Perot. The instrument will be one of the first-light instruments of the 4m aperture Daniel K. Inoue Solar Telescope (DKIST). To provide a field of view of 1 arcmin and a spectral resolution λ/Δλ of about 100.000, the required free aperture of the Fabry Perot is 250mm. The high reflectivity coatings for the Etalon plates need to meet the specifications for the reflectivity over the entire wavelength range and preserve the plate figure specifications of better λ/300, and a micro roughness of < 0.4 nm rms. Coated surfaces with similar specifications have successfully been made for reflecting mirrors on thick substrates but not for larger format Fabry-Perot systems. Ion Beam Sputtering (IBS) based coatings provide stable, homogeneous, and smooth coatings. But IBS coatings also introduce stresses to the substrate that influence the plate figure in our case at the nm level. In a joint effort with an industry partner and a French CNRS research laboratory, we developed and tested processes on small and full size substrates, to provide coated Etalon plates to the required specifications. Zygo Extreme Precision Optics, Richmond, CA, USA, is polishing and figuring the substrates, doing the metrology and FE analysis. LMA (Laboratoire Matériaux Avancés, Lyon, France) is designing and making the IBS coatings and investigating the detailed behavior of the coatings and related processes. Both partners provide experience from manufacturing coated plane optics for gravitational wave detection experiments and EUV optics. The Kiepenheuer-Institut für Sonnenphysik, Freiburg, Germany is designing and building the VTF instrument and is leading the coating development. We present the characteristics of the coatings and the substrate processing concept, as well as results from tests on sample size and from full size substrate processing. We demonstrate that the tight specifications for a single Etalon can be reached.
In order to enhance the final performances of complex optical systems it is required to limit the overall wasted reflected light coming from all the different surfaces involved. Ultra-low-reflectance coating becomes even a crucial point for high sensitivity experiments such as gravitational wave detectors where surfaces must have a reflectance lower than 100 ppm. Some tenths of percent is a common value for Anti-Reflective (AR) coating but reflectance below 100 ppm is trickier to achieve. The coating design sensitivity with respect to thickness errors or refractive index error can lead quickly to noncompliant reflectance.
When an AR coating has failed it is very difficult to recover the low reflectance. In theory adding one or two layers could correct the reflectance but it requires knowing exactly the actual coated stack. For large optics (diameter up to 500 mm), we developed a new technique based on reflectance measurements with different polarizations and incidence angles at one wavelength. The measurements were performed in s-polarized and p-polarized light to discriminate between several solutions. Then a correction based on one or two layers is computed in order to decrease the reflectance. The efficiency of this method is demonstrated in the case of a four-layer AR coating designed for zero reflectance at 1064nm and coated onto a 350mm diameter and 200mm thick substrate. The reflectance has decreased from 500ppm to 26ppm thanks to a correcting bilayer.
The LSST design foresees the use of six wide-band large optical filters that can alternatively be moved in front
of the CCD camera. Each of the six filters has a different band-pass covering all the wavelengths from 300 nm to 1200
nm. The way to achieve this is to coat an optimized optical thin films stack on a filter substrate. Each filter requires a
specific design using specific appropriate materials. The main characteristics of these filters, that constitute a real
technological challenge, are: their relatively large size - their radii of curvature (about 5.6 m) that represent a sagitta of
12,5 mm that increases the uniformity complexity, the large rejection band requirements with transmission lower than
0.01 % out of the band and a transmission of 95 % over the band-pass. This paper proposes to show the problematic and
the results obtained at LMA (Laboratoire des Matériaux Avancés-FRANCE) to the purpose of realizing these filters
using the IBS (Ion Beam Sputtering) deposition technique. The results obtained with High-Pass/Low-Pass structures will
be presented. Experimental results will be shown concerning the R-band filter (552-691 nm). An overview of the work to
be done to realize transmittance map over large filters will be given.
The second generation of gravitational wave detectors will aim at improving by an order of magnitude
their sensitivity versus the present ones (LIGO and VIRGO). These detectors are based on long-baseline Michelson
interferometer with high finesse Fabry-Perot cavity in the arms and have strong requirements on the mirrors quality.
These large low-loss mirrors (340 mm in diameter, 200 mm thick) must have a near perfect flatness. The coating process
shall not add surface figure Zernike terms higher than second order with amplitude <0.5 nm over the central 160 mm
diameter. The limits for absorption and scattering losses are respectively 0.5 and 5 ppm. For each cavity the maximum
loss budget due to the surface figure error should be smaller than 50 ppm. Moreover the transmission matching between
the two inputs mirrors must be better than 99%.
We describe the different configurations that were explored in order to respect all these requirements. Coatings are done
The two first configurations based on a single rotation motion combined or not with uniformity masks allow to obtain
coating thickness uniformity around 0.2 % rms on 160 mm diameter. But this is not sufficient to meet all the
A planetary motion completed by masking technique has been studied. With simulated values the loss cavity is below 20
ppm, better than the requirements. First experimental results obtained with the planetary system will be presented.
The Large Synoptic Survey Telescope (LSST) uses a novel, three-mirror, telescope design feeding a camera system that
includes a set of broad-band filters and three refractive corrector lenses to produce a flat field at the focal plane with a
wide field of view. Optical design of the camera lenses and filters is integrated in with the optical design of telescope
mirrors to optimize performance. We discuss the rationale for the LSST camera optics design, describe the methodology
for fabricating, coating, mounting and testing the lenses and filters, and present the results of detailed analyses
demonstrating that the camera optics will meet their performance goals.
Gravitational wave detectors such as Virgo and LIGO use long-baseline Michelson interferometers with high
finesse Fabry-Perrot cavity in the arms. The symmetry of these cavities is essential to prevent the interferometer
from sensitivity to laser fluctuations. For this purpose the difference between the transmissions of the two input
mirrors has to be minimized. Advanced LIGO, the upgrade of LIGO, plans a transmission matching between the
two input mirrors as high as 99%. A small deviation in the process fabrication from run to run might induce
transmission mismatch larger than 1%. Consequently, the two input mirrors have to be coated during the same
coating run. That requires ability to deposit the reflective coating, based on a stack of titanium doped tantala
(Ti:Ta2O5) layers and silica layers, uniformly over a 800 mm diameter aperture. This paper presents the study to
improve the thickness uniformity of a reflective coating and the preliminary results achieved on two Ø350mm
substrates coated in the run.
Through its participation to European programs, SAGEM has worked on the design and manufacturing of normal
incidence collectors for EUV sources. By opposition to grazing incidence, normal incidence collectors are expected to
collect more light with a simpler and cheaper design. Designs are presented for the two current types of existing sources:
Discharge Produced Plasma (DPP) and Laser Produced Plasma (LPP). Collection efficiency is calculated in both cases. It
is shown that these collectors can achieve about 10 % efficiency for DPP sources and 40 % for LPP sources. SAGEM
works on the collectors manufacturability are also presented, including polishing, coating and cooling. The feasibility of
polishing has been demonstrated with a roughness better than 2 angstroms obtained on several materials (glass, silicon,
Silicon Carbide, metals...). SAGEM is currently working with the Institut d'Optique and the Laboratoire des Materiaux
Avancés on the design and the process of EUV coatings for large mirrors. Lastly, SAGEM has studied the design and
feasibility of an efficient thermal control, based on a liquid cooling through slim channels machined close to the optical
The coating deposition on large optical components (diameter 350 mm) has required the development of new metrology tools at 1064 nm. To give realistic values of the optical performances, the whole surface of the component needs to be scanned. Our scatterometer (commercial system) has been upgraded to support large and heavy samples. The other metrology tools are prototypes we have developed. We can mention the absorption (photothermal effect) and birefringence bench, a control interferometer equipped with an original stitching option, the optical profilometer (RMS roughness and small defect measurements). A detailed description of these metrology benches will be exposed. Their sensitivity, accuracy and capability to map the optical properties of substrates or mirrors will be discussed. We will
describe the recent developments: the stitching option adapted to the Micromap profilometer to measure the RMS roughness on larger area (exploration of a new spatial frequency domain), the accurate bulk absorption calibration.