Secrets of adaptive optics: how astronomers observe the starry sky.

The experiment that breathed new life into the science of space was not conducted at a famous observatory or on a giant telescope. Specialists learned about it from the article "Successful Tests of Adaptive Optics," published in the astronomical journal The Messenger in 1989. The results presented were from tests of the electro-optical system Come-On, designed to correct atmospheric distortions of light from cosmic sources. These tests took place from October 12 to 23 on the 152-cm reflector at the French observatory OHP (Observatoire de Haute-Province). The system performed so well that the authors began the article with the assertion that "the long-held dream of astronomers working with ground-based telescopes has finally come true thanks to the creation of new optical observation techniques — adaptive optics."

Just a few years later, adaptive optics (AO) systems started to be installed on larger instruments. In 1993, they equipped the 360-cm telescope of the European Southern Observatory (ESO) in Chile, soon after a similar instrument in Hawaii, and then on 8−10-meter telescopes. Adaptive optics allowed ground-based instruments to observe celestial bodies in visible light with a resolution that was previously the domain of the Hubble Space Telescope, and even higher in infrared rays. For instance, in the very useful astronomy band of the near-infrared region at a wavelength of 1 µm, Hubble achieves a resolution of 110 angular milliseconds, while the 8-meter telescopes of ESO reach up to 30 milliseconds.

In fact, when French astronomers were testing their AO system, similar devices already existed in the United States. However, they were not created for the needs of astronomy. The Pentagon was the client for these developments.

When observing two stars in a telescope that are very close to each other in the sky, their images will merge into a single glowing point. The minimum angular distance between such stars, determined by the wave nature of light (the diffraction limit), is the resolution of the instrument. This resolution is directly proportional to the wavelength of light and inversely proportional to the diameter (aperture) of the telescope. For a three-meter reflector in green light observations, this limit is about 40 angular milliseconds, while for a 10-meter telescope, it is just over 10 milliseconds (at this angle, a small coin can be seen from a distance of 2000 km).

However, these estimates are only valid for observations in a vacuum. In the Earth's atmosphere, local turbulence constantly arises, changing the density and temperature of the air several hundred times per second, and consequently, its refractive index. Therefore, in the atmosphere, the front of the light wave from a cosmic source inevitably spreads out. As a result, the actual resolution of ordinary telescopes, at best, is 0.5−1 arc second, which is significantly below the diffraction limit.

Imagine a device that analyzes the light waves passing through the telescope hundreds of times per second to detect traces of atmospheric turbulence and, based on this data, alters the shape of a deformable mirror placed at the telescope's focus to neutralize atmospheric disturbances and ideally make the image of the object "vacuum-like." In this case, the resolution of the telescope is limited solely by the diffraction limit.

However, there is one nuance. Usually, the light from distant stars and galaxies is too weak for reliable reconstruction of the wave front. It is a different matter if there is a bright source near the observed object, whose rays travel to the telescope almost along the same path — these can be used to read atmospheric disturbances. This scheme (in a somewhat reduced form) was tested by French astronomers in 1989. They selected several bright stars (Deneb, Capella, and others) and, using adaptive optics, indeed improved the quality of their images when observing in infrared light. Soon, such systems using guide stars from the Earth's sky began to be employed on large telescopes for real observations.

However, there are not many bright stars in the Earth's sky, so this method is suitable for observing only about 10% of the celestial sphere. But if nature has not created a suitable star in the needed location, an artificial star can be created — by using a laser to stimulate the atmosphere at a high altitude, which will become a reference light source for the compensating system.

This method was proposed in 1985 by French astronomers Reno Fua and Antoine Labeyrie. Around the same time, their colleagues in the U.S., Edward Kibblewhite and Laird Thomson, reached similar conclusions. In the mid-1990s, laser emitters paired with AO equipment appeared on medium-sized telescopes at the Lick Observatory in the U.S. and at the Calar Alto Observatory in Spain. However, this technique took about ten years to find application on 8−10-meter telescopes.

The history of adaptive optics has both a clear and a secret side. In January 1958, the Pentagon established a new structure, the Advanced Research Projects Agency (ARPA, now DARPA), responsible for developing technologies for new generations of weaponry. This agency played a pivotal role in the creation of adaptive optics: telescopes insensitive to atmospheric disturbances with the highest possible resolution were needed to observe Soviet orbital devices, and there was also a future consideration for creating laser weapons to destroy ballistic missiles.

In the mid-1960s, ARPA launched a program to study atmospheric disturbances and the interaction of laser radiation with air. This was conducted at the RADC (Rome Air Development Center) research center, located at Griffiss Air Force Base in New York. Powerful searchlights mounted on bombers flying over the testing ground were used as a reference light source, and this was so impressive that frightened residents would sometimes call the police!

In the spring of 1973, ARPA and RADC contracted the private corporation Itec Optical Systems to participate in the development of devices that compensate for light scattering due to atmospheric disturbances under the RTAC (Real-Time Atmospheric Compensation) program. Itec employees created all three main components of AO — an interferometer for analyzing wave front disturbances, a deformable mirror for correcting them, and a control system. Their first two-inch diameter mirror was made of glass coated with an aluminum reflective film. Piezoelectric actuators (21 in total) were built into the support plate, capable of contracting and extending by 10 µm in response to electrical impulses. Initial laboratory tests conducted that same year indicated success. The following summer, a new series of tests demonstrated that the experimental equipment could correct a laser beam over distances of several hundred meters.

These purely scientific experiments had not yet been classified. However, in 1975, a classified program called CIS (Compensating Imaging System) was approved to develop AO for the Pentagon's interests. Under this program, more advanced wave front sensors and deformable mirrors with hundreds of actuators were created. This equipment was installed on a 1.6-meter telescope located on the summit of Haleakalā on the Hawaiian island of Maui. In June 1982, it was used to obtain photographs of an artificial Earth satellite for the first time with acceptable quality.

Although experiments in Maui continued for several more years, the development center moved to a special area of Kirtland Air Force Base in New Mexico, to the secret Sandia Optical Range (SOR), where work on laser weapons had been ongoing for a long time. In 1983, a group led by Robert Fugate began experiments to study laser scanning of atmospheric inhomogeneities. This idea was proposed in 1981 by American physicist Julius Fainleib, and now it needed to be tested in practice. Fainleib suggested using elastic (Rayleigh) scattering of light quanta on atmospheric inhomogeneities in AO systems. Some of the scattered photons return to the point from which they left, and a characteristic glow almost like a point source occurs in the corresponding area of the sky — an artificial star. Fugate and colleagues registered the distortions of the wave front of the reflected radiation on its way to Earth and compared them with similar disturbances of starlight coming from the same area of the sky. The disturbances turned out to be almost identical, confirming the feasibility of using lasers for AO tasks.

These measurements did not require complex optics — simple mirror systems sufficed. However, for more reliable results, they needed to be repeated on a good telescope, which was installed at S