SPIE Membership Get updates from SPIE Newsroom
  • Newsroom Home
  • Astronomy
  • Biomedical Optics & Medical Imaging
  • Defense & Security
  • Electronic Imaging & Signal Processing
  • Illumination & Displays
  • Lasers & Sources
  • Micro/Nano Lithography
  • Nanotechnology
  • Optical Design & Engineering
  • Optoelectronics & Communications
  • Remote Sensing
  • Sensing & Measurement
  • Solar & Alternative Energy
  • Sign up for Newsroom E-Alerts
  • Information for:
SPIE Photonics West 2019 | Call for Papers

2018 SPIE Optics + Photonics | Register Today



Print PageEmail Page


New telescopes herald a golden age of astronomy

From OE Reports Number 195 - March 2000
29 March 2000, SPIE Newsroom. DOI: 10.1117/2.5200003.0005

"We are all living in the gutter, but some of us are looking at the stars." Oscar Wilde, Lady Windermere's Fan

Figure 1. The new class of 8-m telescopes coming on line. Top to bottom: the Very Large Telescope atop Cerro Paranal in Chile; Gemini North on Mauna Kea, Hawaii; and Subaru, also atop Mauna Kea. Images courtesy of the respective observatories.

For astronomers, space and time are inextricably linked. The farther you see into the universe the further back in time you go to approach that moment of creation, the Big Bang. The bigger the redshift of an object, such as a galaxy, the farther it is from us in both space and time, giving us a glimpse of the early universe.

Three years ago, at Keck headquarters, I watched as Caltech astronomer Chuck Steidel and graduate student Kurt Adelberger studied distant galaxies with redshifts over 3.0, representing the age of the universe 85 percent back in time. They had found a wall of galaxies at redshift z=3.1 and smaller peaks at other redshifts, prompting some theorists to believe these density peaks could originate from cosmic clumping caused by oscillations rippling through the hot, ionized gas of the early universe.

Steidel's work also has significance for the evolution of the universe and the question of dark matter. "We don't see enough objects emitting light to account for all the mass that we infer from gravity," Adelberger said. "In principle, it is possible to learn about the mass of dark matter particles by measuring the size of the structures in the young universe."

These kinds of glimpses into the science provide the raison d'etre of large-aperture telescopes. Questions about the universe will be answered, and undreamed-of phenomena will be discovered because of this new age of telescopes, reflected by the 10-m Keck I and II, the 10-m Hobby-Eberly, and the newly "turned on" 8-m aperture-size telescopes (Figure 1) such as Gemini, Subaru, the European Southern Observatory's (ESO) Very Large Telescope (VLT), and, under construction, the Large Binocular Telescope in Arizona.

"If you want to analyze faint objects, size matters," said Kaz Sekiguchi, General Manager of the 8.3-m Subaru Telescope atop Mauna Kea. And while it is true the Hubble, at 2.4 m, provides remarkable clarity, it cannot capture the faint spectra its larger, earthbound cousins can. And with adaptive optics (Figure 2), which uses wavefront sensors and deformable mirrors to cancel out the aberrations caused by the atmosphere, groundbased telescope images will approach the quality of the Hubble. "We can get about the same image size," Sekiguchi said. "And, most of all, it is much cheaper."

Figure 2. (a) An image of Neptune taken by the Keck telescope without adaptive optics. (b) Photo of Neptune taken with adaptive optics. Images were obtained in broadband J and H (1.25 and 1.65 µm -- blue and red, respectively) and in the methane absorption band at 1.17 µm in the J band (green), and then combined. Neptune itself was used as a guide star, despite its large diameter (2.3 arcsecond).

Subaru, on Mauna Kea, cost $400 million to build and is operated by the National Astronomical Observatory of Japan. The Gemini Observatory, which has two telescopes, Gemini North atop Mauna Kea and Gemini South atop Cerro Pachón in Chile, cost $184 million. Gemini has seven partner countries -- the United States, Great Britain, Canada, Chile, Argentina, Brazil, and Australia. The VLT, atop Cerro Paranal in the Atacama Desert in Chile, cost $600 million. It is run by ESO, whose member nations are Belgium, Denmark, Germany, France, Italy, the Netherlands, Sweden, and Switzerland. ESO was formed in 1962 to provide a major observatory for European scientists to study the southern sky.

Figure 3. Every 30 seconds, 150 axial and 64 lateral actuators on the back of the VLT primary mirror bend and hold it in the shape of a hyperboloid.

The telescopes of the three observatories are Ritchey-Chretien in design. Their primary mirrors were formed from low thermal expansion coefficient glass, Ultra Low Expansion glass from Corning (Canton, New York) in the case of Subaru and Gemini and Zerodur from Schott Glaswerke (Mainz, Germany) for the VLT. The shape of the primary mirror is hyperboloid and shaped by actuators, the number varying for each observatory. There are 261 actuators for Subaru, 150 axial and 64 lateral actuators for the VLT (Figure 3), and 120 axial, 60 lateral, and 24 stabilizing actuators for the Gemini. The diameter of the Gemini is 8.1 m, 8.2 m for the VLT, and 8.3 m for the Subaru. The two primary mirrors for the Large Binocular Array in Arizona will be 8.4 m.

The primary mirror blanks for the Gemini and Subaru were fused, ground, heated, and sagged onto a curved form to assume the final meniscus form. The VLT mirror blank was spun cast. In forming a VLT mirror blank, 45 tons of glassy Zerodur were poured into a concave mold, which was then put on a rotating platform where it was spun until it became solid. When the temperature decreased to 800 deg. C with the blank maintaining its meniscus shape, it was put into an annealing furnace where it gradually cooled over three months. The first surface to be removed was the thin crystalline layer that built up during annealing. Both the concave and convex sides of the blank were machined to specifications. Then the blank was returned to the furnace for the ceramization cycle, which lasted about nine months. Once the blank was rough figured, the back of the blank was set up for the actuators. It was then sent on to final figuring down to sub-µm accuracy, then polished.

The VLT has four telescopes that will search the southern sky. Each telescope can be used independently, or the four fixed telescopes can be combined with three moveable auxiliary 1.8-m telescopes into an interferometer. As for using the four telescopes together as a "light bucket" -- essentially a 16-m telescope -- for the incoherent combination of light, Head of Optical Instrumentation at ESO Sandro D'Odorico said, "We do not foresee for the time being implementing it." The large number of mirrors and optics required would make the system too lossy, thus ineffective at combining the light in a meaningful way.

As an interferometer, the light is coherently combined at a central laboratory. Figure 4 shows the process for two telescopes. A star at infinity illuminates the apertures of the two telescopes with a plane wave that is guided through the Coudé Optical Trains into the Delay Line Tunnel. The light signals arriving at telescopes 1 and 2 are compensated by the delay lines so that the beams have zero Optical Path Difference (OPD) when they interfere on the detector in the VLTI laboratory.

Figure 4. Schematic showing the VLT interferometer for two telescopes. Images of the same object are captured by two telescopes and brought to the two Coudé foci, where a delay line is introduced into the pathlength of one so that the two beams recombine with zero optical path difference.

The field of view of the VLTI is 2 arcsec. However, a dual-feed facility will allow picking two stars at the Coudé focus of the telescopes, each in a 2-arcsec field of view and separated by up to 1 arcmin. This is useful for astrometry, allowing you to cophase the telescopes with a bright reference object within a radius of about 30 arcseconds of a fainter science target that would otherwise be too faint for direct detection. One integrates on the science object for a long time to get a good signal-to-noise ratio, which should enable the detection of the wobble in a star caused by a planet the size of Uranus.

Using different baselines (in length and orientation) allows reconstructing an image with the angular resolution of a single telescope with a diameter equal to the longest baseline, 130 m for the four 8-m telescopes and 200 m for the three 1.8-m auxiliary telescopes. "This makes the VLTI array one of the most powerful tools available in this decade for very high resolution astronomy," said ESO VLT Interferometer Project Scientist Francesco Paresce.

Figure 5. This is a three-color composite image of RCW38 taken by the ESO VLT Infrared Spectrometer And Array Camera (ISAAC) mounted on the first 8.2-m unit telescope (UT1). It was obtained through three near-infrared filters. This is a region in the Milky Way at a distance of about 5,000 light years. These stars, recently formed in clouds of gas and dust, are still heavily obscured and cannot be observed in the visible part of the spectrum, but are easily seen at infrared wavelengths where the obscuration is substantially lower. The diffuse radiation is a mixture of starlight, scattered by the dust and gas in the area, and atomic and molecular hydrogen line emission.

To minimize the thermal signatures inside the domes, the inside temperature must be the same as the outside temperature. The different designs of the enclosures and structures indicate the various ways to achieve this with minimal turbulence. "The Subaru and the ESO VLT telescopes have a corotating dome that does not allow independent rotation of the telescope inside the dome," said Gemini Optics Manager Larry Stepp. The Keck and Gemini enclosures, on the other hand, are more traditional. They are spherical in design and the telescopes can rotate independently of the dome.

"All of the infrared is coming not just from the background of the sky, but the building, telescope structure, and so on," Sekiguchi said. Running fluid dynamics tests of air flow through many different designs, Subaru settled on the cylindrical model because, according to Sekiguchi, "The classical dome shape makes some air turbulence." The shape and many side vents, controlled by louvers, give a steady airflow with minimal turbulence in front of the telescope and inside the rest of the dome.

Gemini also has large openings on the side for natural ventilation (Figure 1), which readily brings the facility and telescope into equilibrium with the nighttime air temperature.

Minimizing the background thermal signatures enables the telescopes and instruments to observe not only in the visible, but in the near-to-mid infrared as well. Interest in these wavelengths has grown because there are large redshifts in that region, there is less distortion by the atmosphere, IR radiation penetrates gas and dust clouds in galaxies and nebulae better than visible light, and large array detectors have been developed in recent years. For example, at Subaru, the Coronagraphic Imager with Adaptive Optics is an infrared coronagraph, which can cover the stellar image to detect the spectra of very faint companions such as brown dwarfs, which are intermediate in mass between stars and planets.

The new observatories are also actively pursuing adaptive optics to help them achieve better image quality. The atmosphere deforms plane waves, or light from a star -- the science object. Adaptive optics aims to nullify the effects of atmospheric turbulence on the light reaching the telescope (see Interview with John Hardy,OE Reports, December 1994, and Interview with Peter Wizinowich, OE Reports, July 1997). In a nutshell, light from a natural or laser guidestar is used as a reference to deform a mirror to cancel the atmospheric distortions, which, to first order, are localized tip/tilts across the guidestar wavefront. The actuators on a deformable mirror are then adjusted to null all of these tip/tilts and flatten the guidestar wavefront.

Usually, this reading of the atmospheric turbulence covers only a minuscule area of the sky, typically about 20 arcseconds (1/200 of a degree) across. Gemini scientists are proposing multiconjugate adaptive optics (MCAO) to expand the sky coverage. "The multi-conjugate systems will use multiple wavefront sensors to measure turbulence over a range of directions," said "Brent Ellerbroek, Adaptive Optics Program Manager for the Gemini Observatory. Using algorithms is a little bit like tomographic techniques used with CAT scans: we can estimate in all three directions, instead of one direction. With multiple deformable mirrors, we can implement a correction, which will give us much more uniform performance over a wide range." Ellerbroek estimates an order of magnitude improvement over standard AO.

The plan is to use five relatively bright laser guidestars, arranged either as an X or cross (Figure 5), with center-to-corner separation of 1/2 to 3/4 arcminute. High-power lasers (10 W) are aimed at the sodium layer, 90 ± 10 km high in the atmosphere. At 589 nm, the laser light is absorbed by the sodium D2 line, whereupon the sodium atom fluoresces to become the laser guidestar used for adaptive optics.

Figure 6. (a) Laser guidestar geometry for MCAO at Gemini-South. Five laser guidestars are generated at an altitude of 90 km in the mesospheric sodium layer. Three deformable mirrors, optically conjugate to ranges of 0, 4.5, and 9.0 km, correct for atmospheric turbulence across a square field of view with a 1.6-arcminute diagonal, approximately three times the diameter of the field that can be corrected using a single guidestar and deformable mirror. (b) The beamprints at 9 km. The science beam envelope is the union of all beamprints from all science observations over a 1.6-arcminute field of view. The on-axis science beamprint is from a star at infinity. The five laser guidestars form the five circular beamprints (dashed lines), whose centers are displaced in x and y with respect to each other to form the X of (a). The beamprints of the laser guidestars are used to fill the volume at this altitude to measure the turbulence.

There are still issues to be addressed for MCAO. Fluorescing sodium atoms 90 km high are not exactly pinpoints of light. "This means that the laser guidestar wavefront is not a plane wave, but a spherical wave," said Celine d'Orgeville, Laser System Engineer at Gemini. To achieve the best results for adaptive optics, i.e., smallest spot size, atmospheric conditions need to be good because the laser beam is distorted on the way up. The laser needs to be as close to diffraction limited as possible. Moreover, with a natural guidestar the turbulence volume is a cylinder whereas with a laser guidestar the turbulence profile is cone-like. This is called the cone effect or focal anisoplanatism and has a large detrimental effect when telescope apertures become too large (30-100 m in the future). With MCAO, because the laser guidestar beamprints fill up the turbulence volume more (Figure 5b), the problem is solved. In effect, they fill the same cylindrical volume as would be seen from a natural guidestar.

With no off-the-shelf 589-nm, 10-W laser available, future directions for the Gemini MCAO program include increasing the laser power, developing reliable, low-maintenance, fully automated laser systems, and driving down costs. "We are funding multiple laser R&D programs to do risk-reduction demonstrations of some promising laser technologies. By fostering competition, we expect the unit cost to go down," d'Orgeville said.

These new telescopes with their new technology promise a golden age of astronomy. They will provide clear pictures of a cataclysmic universe, and perhaps raise more questions for theorists, as they probe deeper into the universe than ever before.

Sandra Faber, UC Santa Cruz astronomer, once wrote me, "I think understanding the origin of the universe is the boldest and most noble quest that science can undertake, and I am proud to be part of it."

The scientists and engineers who built these new telescopes can be proud, too.

Frederick Su

Frederick Su is a freelance writer based in Bellingham, WA. Web: www.bytewrite.com.