Multi-laser-guided adaptive optics for the Large Binocular Telescope

We describe the conceptual design of an advanced laser guide star facility (LGSF) for the Large Binocular Telescope (LBT), to be built in collaboration with the LBT’s international partners. The highest priority goal for the facility is the correction of ground-layer turbulence, providing partial seeing compensation in the near IR bands over a 4 field. In the H band, GLAO is projected to improve the median seeing from 0.55 to 0.2 . The new facility will build on the LBT’s natural guide star AO system, integrated into the telescope with correction by adaptive secondary mirrors, and will draw on Arizona’s experience in the construction of the first multi-laser adaptive optics (AO) system at the 6.5 m MMT. The LGSF will use four Rayleigh beacons at 532 nm, projected to an altitude of 25 km, on each of the two 8.4 m component telescopes. Initial use of the system for ground layer correction will deliver image quality well matched to the LBT’s two LUCIFER near IR instruments. They will be used for direct imaging over a 4 4 field and will offer a unique capability in high resolution multi-object spectroscopy. The LGSF is designed to include long-term upgrade paths. Coherent imaging at the combined focus of the two apertures will be exploited by the LBT Interferometer in the thermal IR. Using the same launch optics, an axial sodium or Rayleigh beacon can be added to each constellation, for tomographic wavefront reconstruction and diffraction limited imaging over the usual isoplanatic patch. In the longer term, a second DM conjugated to high altitude is foreseen for the LBT’s LINC-NIRVANA instrument, which would extend the coherent diffraction-limited field to an arcminute in diameter with multi-conjugate AO. 1. THE LBT AND ITS INSTRUMENTATION The LBT, shown in a recent photograph in Figure 1, comprises two 8.4 m primary mirrors on a common mount. The telescope is now operating with a prime focus camera as the first-light science instrument. Several instruments will be deployed at the telescope over the next two years that can operate autonomously, but will also benefit from the application of adaptive optics (AO). Of primary importance for early application of the LGSF is a pair of instruments called LUCIFER, one of which will attach to each of the LBT’s unit telescopes. The two copies are being built by a consortium of five German institutes as part of Germany’s contribution to the LBT; one unit is shown in final assembly in Figure 2. They are scheduled for first light in early 2008 and early 2009. Each is a near IR spectrograph and imager working in the wavelength range from 0.9 to 2.5 μm. They are multi-mode instruments intended for seeing limited as well as diffraction limited operation. The main features are: • Direct imaging over a 4 4 field of view (FOV) in the seeing limit with 0.12 pixels. • Multi-object spectroscopy with cold slit masks. A 23-mask cassette is provided. • Long-slit spectroscopy with 4 slit and resolution up to 10,000 in the seeing limit and up to 40,000 in the diffraction limit. Two gratings are provided with a third slot available. • Diffraction-limited imaging over a 30 30 FOV with 0.015 pixels. The detectors are each 2048 2048 Hawaii 2 arrays. They are optimized differently, with detector quantum efficiencies as shown in Table 1. When coupled with AO, LUCIFER will offer a unique capability on 8-m class telescopes: high resolution multi-object spectroscopy in the near IR, which has particular application in cosmology to the study of high redshift galaxies. Figure 1. The Large Binocular Telescope on Mt. Graham, Arizona has two 8.4 m primary mirrors on a common mount. Astronomical Adaptive Optics Systems and Applications III, edited by Robert K. Tyson, Michael Lloyd-Hart Proc. of SPIE Vol. 6691, 66910O, (2007) · 0277-786X/07/$18 · doi: 10.1117/12.734821 Proc. of SPIE Vol. 6691 66910O-1 Downloaded from SPIE Digital Library on 27 Mar 2010 to 128.196.208.1. Terms of Use: http://spiedl.org/terms The laser-guided AO system will also serve to sharpen the images seen by the LBT Interferometer. With 23 m baseline, the LBT is now the largest optical telescope in the world on a single mount, an essential factor in providing a large FOV at the coherent combined focus. LBTI, under construction at Arizona and scheduled to come on line in 2009, will be the first instrument to take full advantage of the resolution offered by coherent combination of the beams from the two unit telescopes. Operating in the thermal IR from 5 to 20 μm, LBTI will take advantage of the laser AO capability to probe regions of scientific interest where no bright natural star is available. These will typically be extragalactic sources, for instance probing the stellar dynamics and the spectral energy distribution of the disk around the central black hole in M31, or, more locally, heavily obscured regions such as the center of our own Galaxy. With AO in operation, the LBT will place just four warm surfaces between the sky and the LBTI dewar. For operation at these thermal wavelengths it will therefore be more sensitive than either the Keck or VLT interferometers. Later on, the high-resolution interferometric capability will be extended to the near IR with the addition of LINC-NIRVANA, under construction at the Max-Planck Institute for Astronomy in Heidelberg as a collaboration between LBT’s German and Italian partners. Expected to be operational in 2010, LINC-NIRVANA will offer a 20 field of view with resolution as high as 10 milliarcsec. To give an example of its application, at this resolution, and with the sensitivity of LBT’s aperture, equivalent to a 12 m single aperture, stellar populations in galaxies at 5 – 20 Mpc will be resolved. For the first time, individual stars in giant elliptical galaxies will be within reach, allowing their star formation history to be investigated directly. 1.1. Adaptive optics at the LBT The LBT was designed from the outset to include AO as an integral part of the telescope, and to that end the telescope is being equipped with a pair of adaptive secondary mirrors (ASM) of the same construction as the ASM built for the 6.5 m MMT. Natural guide star (NGS) wavefront sensors are nearing completion. First light with the first ASM is anticipated in late Spring 2008 with the second ASM to follow one year later. Each has 672 actuators, projecting to 29 cm spacing on the primaries, and a response time of 0.5 ms. The first generation of AO instruments, LUCIFER, LBTI, and LINCNIRVANA, will operate in the near and thermal IR, but the capacity will be in place to support a second generation that will push AO correction down to visible wavelengths. High-order AO compensation is now de rigueur for large general-purpose astronomical telescopes, with a general expectation that over limited fields of view, diffraction-limited imaging will be available in at least the near IR J, H, and K bands. The system to be implemented at the LBT relies on an ASM to ensure the highest optical efficiency and more importantly the lowest emissivity for work that extends into the thermal IR. The first-light AO system will rely entirely on NGS and therefore will be limited in sky coverage, particularly for work at high redshift where many objects of interest are to be found in otherwise blank regions of the sky, and in highly obscured regions such as stellar nurseries. In such regions, no sufficiently bright star will be found to serve as a reference beacon for AO wavefront correction. A laser-guided AO system is therefore essential to realize the full scientific potential of the LBT, and such a system is now planned by three major partners in the LBT consortium: the University of Arizona, the Max Planck Institut für extraterrestrische Physik in Garching, and the Osservatorio Astrofisico di Arcetri in Florence. The design, described in detail in the following sections, builds on pioneering work with multiple Rayleigh laser guide stars (RLGS) at the 6.5 m MMT. This approach offers a number of compelling advantages. In the first instance, the cost of a system which relies on a number of lasers that are cheaply available from commercial vendors is substantially less than a system based on even a single sodium LGS. The cost of a 10 W sodium laser alone, such as the one recently installed on Gemini North, would exceed $5M, and it is not clear that the LBT could operate satisfactorily with a single center-mounted sodium LGS. The effects of spot elongation (which would be comparable to the Keck II telescope) and off-axis anisoplanatism (which would be substantially worse) would combine to reduce the achievable Strehl ratio. A better system would deploy two sodium LGS, projected from behind the two ASMs, but the cost of such a system would be much higher. Figure 2. The first of the two LUCIFER imager/ spectrographs near completion. Waveband: J H K LUCIFER 1 0.33 0.75 0.73 LUCIFER 2 0.53 0.58 0.55 Table 1. LUCIFER detector quantum efficiencies Proc. of SPIE Vol. 6691 66910O-2 Downloaded from SPIE Digital Library on 27 Mar 2010 to 128.196.208.1. Terms of Use: http://spiedl.org/terms MED AN CN2 PROFILE OF ALL PROFILES — median profile +/— 25% variation 1o 1o_16 cri2 value[ m2"31 a) 0 00 (0 C) C) C) 0 0 0 -c C) -c 10 15 In addition to lower cost, the application of multiple RLGS offers scientific benefits. Detailed site studies with SCIDAR have confirmed that the LBT site is excellent, with median seeing of 0.67 , and that it manifests a boundary layer of turbulence which frequently accounts for the majority of the seeing; Figure 3 shows the median of the profiles in this study. Removing the effect of this layer would offer image resolution in the near IR bands better than 0.2 over a field of ~5 as a matter of routine. The technique to do this is called ground-layer adaptive optics (GLAO), and relies on measurements of the atmospheric wavefront aberration from a number of beacons placed around the corrected FOV. Wavefront sensing for GLAO is most readily done with a constellation of RLGS. The average of the signals from these beacons yields an estimate of the low-lying aberration which is common to them all. Wavefront correction is best done by a deformable mirror (DM) that is optically conjugated to the ground layer. The LBT is ideally suited for GLAO in this sense since the Gregorian configuration of its ASMs places the conjugate height of the wavefront correction in the boundary layer itself at a height of 100 m. The deployment of multiple RLGS also offers a path to diffraction-limited imaging through tomographic wavefront sensing. In this technique, the signals from the beacons are not simply averaged together, but analyzed to give an instantaneous view of the threedimensional structure of the turbulent aberration. By integrating through this volume along any chosen line of sight, the required wavefront correction can be calculated and applied to one or more DMs that are optically conjugated to layers of particularly strong turbulence in the atmosphere. This use of multiple DMs as well as multiple laser beacons, termed multiconjugate adaptive optics (MCAO), achieves a diffraction-limiited image over an extended FOV, on the order of 1 . The general principle of tomography has been demonstrated on-sky now in three experimental MCAO system. Experiments at the MMT specific to tomographic sensing with multiple RLGS have shown that imaging close to the near-IR diffraction limit can be achieved even with a single DM, with the usual single-conjugate limit imposed by anisoplanatism. Baranec et al. describe the first closed-loop tests with a multi-RLGS system which will lead to a full demonstration of laser tomography. The system planned for the LBT is designed as a phased implementation. In Phase 1, GLAO will be offered with LUCIFER which can already take advantage of the partially corrected image over a 4 field. With modest additional effort, the system can be made to work with LBTI by adding an optical feed to the wavefront sensors (WFS) from the two LBTI ports as well as the LUCIFER ports. The Phase 1 GLAO system designed for partial seeing compensation in the near IR will in fact already be of high enough quality to deliver diffraction-limited performance at the longer thermal wavelengths. Diffraction-limited imaging in the near IR with tomography will be the subject of Phase 2, with additional LGS and WFSs, now drawn into a tighter constellation. The system would then fully exploit LUCIFER’s diffractionlimited imaging and long-slit spectroscopic capabilities. Phase 3 would provide MCAO, already designed as an upgrade path into LINC-NIRVANA, offering the extremely high resolution of the telescope at the coherent focus in the near IR. Ideally, MCAO would be implemented with the addition of a sodium LGS to each half of the telescope. Combined with the lower-altitude RLGS, this hybrid sensing system will require just a single tip-tilt star for full multi-conjugate correction, which would otherwise require three well separated stars. With this in mind, the laser launch optics for the LBT system are being designed to accommodate both 532 nm for the RLGS and 589 nm for the sodium LGS. The LBT, with 23 m baseline in one axis, may be thought of as the first of a new generation of extremely large telescopes (ELT) of 30 m class. The Giant Segmented Mirror Telescope, recommended by the most recent Decadal Survey as the top priority for ground-based astronomy in the US, is the conceptual prototype for a number of new ELT projects in the US and Europe. All will require multi-LGS AO to guarantee all-sky access at full resolution. Single LGS will no longer suffice, being limited by focal anisoplanatism even at the range of a sodium beacon. It is essential then Figure 3. Median, 25, and 75 percentile Cn 2 profiles observed at the LBT during a recent campaign of SCIDAR measurements. H e ig h t a b o v e t e le s c o p e ( k m ) Proc. of SPIE Vol. 6691 66910O-3 Downloaded from SPIE Digital Library on 27 Mar 2010 to 128.196.208.1. Terms of Use: http://spiedl.org/terms