eso9304 — Science Release
Brightest Known Double Quasar Discovered
3 June 1993
Although astronomy is an exact natural science, chance sometimes plays an important role. The recent discovery during an otherwise routine observation of the hitherto brightest known double quasar  would not have been possible without some luck.
What began as a normal inspection of a photographic plate obtained with one of the ESO telescopes, has now resulted in the identification of a unique celestial object by a group of European astronomers . In addition to being of great interest in itself, the new quasar (designated HE 1104-1805 AB) may soon provide observers with an independent estimate of the distance scale of the universe. It is also eminently suited to study the structure and composition of the mysterious absorbing gas clouds in the early universe.
Searching for bright quasars
More than 5000 quasars are now known, but most are seen as rather faint points of light which can only be studied with comparatively large telescopes. Quasars which are bright enough to be observed in detail with different techniques at all wavelengths are quite rare. For this reason, the only way to find such objects is to use wide-angle telescopes to search large parts of the sky.
Since 1990, this group of astronomers has used the ESO i-metre Schmidt telescope at the La Silla observatory for such a quasar search programme, the Hamburg-ESO Bright QSO Survey. The telescope functions as an enormous, wide-field photographic camera and is equipped with a large glass prism, the largest of its kind in the world. In this way, the light from celestial objects in the field of view is separated into different wavelengths (colours) and their spectra are directly registered on the photographic plate. To show the faintest possible objects, the photographic plates are exposed for 75 minutes each. One 30 x 30 cm plate may simultaneously record up to 30,000 spectra of celestial objects which are seen as short, narrow lines.
Next, these plates are scanned with a fast measuring machine (a microphotometer) at the Hamburg Observatory. The data are read into a computer where, after the complete information in the many spectra is evaluated by means of specially developed software. The exact pattern of dark and bright lines in an object's spectrum allows to decide about its nature, for instance whether it is a star in the Milky Way or a distant galaxy. When on rare occasions the presence of a distinctive pattern is seen in a spectrum, the corresponding object is identified as a possible quasar.
However, this is only a rough classification. To be completely certain about their true nature, the "quasar candidates" found this way are subsequently observed individually and in much more detail. This is done by high-dispersion spectroscopy with the l.5-metre and 3.6-metre telescopes on La Silla. If a candidate object is then confirmed as a quasar, its distance is measured by means of the redshift (the shift of the spectral lines towards longer wavelengths due to the quasar's velocity) and information is obtained about its chemical composition and general physical state.
The Hamburg-ESO Bright QSO Survey has been quite successful; in a surveyed sky area that corresponds to less than 4 % of the entire sky, it has already identified more than 200 hitherto unknown, bright quasars whose visual magnitudes are brighter than 17.5.
A new bright double quasar found by chance
In March 1993, Lutz Wisotzki and Thorsten Kohler from the Hamburg Observatory were performing spectroscopic observations of some of the new quasar candidates with the EFOSC instrument (ESO Faint Object Spectrograph and Camera) at the 3.6 m telescope. These observations were made through a narrow slit which permits the light from the quasar as well as that from a small area of the surrounding sky to be recorded simultaneously by the spectrograph. The advantage is that in this way the light from the sky background may later be subtracted to produce a "pure" spectrum of the quasar.
The astronomers turned their telescope towards one of the brightest of the candidate objects. Their notes indicated that it had appeared as a perfectly normal, single object on the photographic plate. And so did it look in the telescope; the astronomers did not notice anything peculiar with this object during their observations. The spectrum of the presumed quasar was duly registered and they soon went on to observe the next object.
It was only some weeks later, during the data reduction in Hamburg that they recognized that the recorded spectrum was double. It was a quasar, but next to a well recorded spectrum of this object, there was another much fainter spectrum which otherwise looked exactly like the first one. The quasar was obviously double!
This unexpected discovery was due to the good chance that the spectrograph slit happened to be aligned in such a way that the light from a hitherto unknown, fainter component (B) was recorded next to that of the brighter one (A). The distance between the two was measured as 3 arcseconds. The spectrum of the B component was not well exposed and did not allow an unambiguous interpretation. It was because of its closeness to the much brighter object A that the faint light from B had not been seen before, neither on the photographic plates, nor at the telescope.
Follow-up observations with the ESO NTT
To understand better the nature of this very unusual object, now given the name HE 1104 1805 AB, Dieter Reimers and Lutz Wisotzki have used the ESO 3.5-metre NTT telescope in the remote control mode via satellite link from the ESO Headquarters in Garching during 2 nights (10 -12 May, 1993). They were able to confirm the double nature of the object and to obtain much more detailed spectroscopic and photometric data.
Here are the latest facts: The two objects of HE 1104-1805 AB form a quasar pair with a separation of 3.0 arcseconds and have very nearly identical spectra. The measured redshifts, z = 2.303, are also virtually equal; they indicate that the pair is moving away from us at a speed of ~250,000 km/sec. This corresponds to a look-back time of 83% of the age of the Universe, i.e. we see them as they were when the Universe was less than one-fifth as old as it is now. Assuming the age of the Universe to be 20,000 million years, the distance to this quasar would be about 16,000 million light-years. The brighter component A has a visual magnitude 16.2, while that of B is 18.0, so A is more than 5 times brighter than B. It is indeed the brightest known object of its kind in the sky and unlike the few other known double quasars, it therefore offers a unique opportunity for even very detailed studies with large telescopes which demand a lot of light.
One or two quasars?
The similarity of the spectra indicates that A and B may in reality be the two images of the same distant quasar, whose light has been bent by an object which is situated between the quasar and us. If so, the quasar light will reach us via two slightly different paths, and two images of different brightness are seen of the same object. This strange phenomenon was first discovered in the late 1970's. It is referred to as gravitational lensing .
Upon closer inspection, however, there appears to be small, but distinct differences between spectra of components A and B. In fact, these differences are exactly what would be expected if component A is amplified by the effect of microlensing due to one or more stars in a distant galaxy being near the line-of-sight to the quasar. It will be easy to test this hypothesis since the difference between the spectra of A and B as observed in May 1993 would then disappear after a few years at the most. But if it does not, then HE 1104-1805 AB must be a genuine pair of quasars and the two images now observed are of two different quasars, situated near each other in space. They may even orbit each other in a binary quasar system. Most astronomers opt for the first case, however.
Should it indeed turn out to be a gravitational lens object, HE 110./-1805 AB may be used to monitor the time delay of the future brightness changes of the two images, i.e. how much later one component will mimic the changes of the other. This measurement, which is quite easy for an object as bright as this, provides a completely independent way of determining the distance scale of the universe, i.e. the Hubble constant Ho. This technique was proposed already in 1964 by Sjur Refsdal, 15 years before the first double image quasar was actually discovered. The predicted time delay between the light variations of the A and B images in this new and bright object is only a few months, that is significantly shorter than for other, fainter double quasars and it can therefore be observed more reliably.
Absorbing clouds in the distant universe
There is another distinct difference between the spectra of the two components which has never before been observed in quasar pairs. In the spectra are seen several strong absorption lines, which show that the quasar light is passing through a metal-rich intergalactic cloud. It lies exactly in the line-of-sight at a distance that corresponds to a redshift of z = 1.66, that is at a look-back time of about 75% of the age of the Universe, or around 1,000 million light-years nearer to us than the quasar. These lines tell us about the thickness and chemical composition of the distant cloud.
Most interestingly, some of the absorption lines from the cloud are only seen in the spectrum of component A, and not in that of component B. This means that the light that forms image A passes through a part of the intergalactic cloud whose density is significantly different from where the B-Iight passes through. From the observed distance in the sky between the two images of 3 arcseconds, the separation of the two light paths at the distance of this cloud is found to be about 60,000 light-years. This gives the first-ever clear size estimate for such a cloud. It is possible that the "lensing" object which causes the bending of the light is also at this distance, since the absorbing cloud has properties known from the disks of galaxies.
The astronomers are now eagerly awaiting to observe HE 1104-1805 AB at high spectral resolution in the ultraviolet spectral region with the Hubble Space Telescope (HST). Because of the high redshift, important absorption lines of the major elements of carbon, nitrogen and oxygen which are emitted in the extreme ultraviolet spectral region that is normally not observable, are here shifted into the observable spectral region. This will make it possible to perform for the first time detailed and quantitative composition and structural studies of an absorbing gas cloud around one or more, still young galaxies in the distant and early universe.
Observations with the IUE (International Ultraviolet Explorer) satellite, performed on April 29 at short notice after discovery of the new object, have confirmed that HE 1104-1805 AB is also very bright in the ultraviolet spectral region and will therefore be a relatively easy target for the HST.
From the ROSAT All Sky Survey, it is also known that the new double quasar is a strong X-ray emitter. It will now be important to find out whether these X-rays are from the quasar itselfor perhaps are due to an intervening cluster of galaxies that is in part responsible for the gravitational lensing effect. Dedicated observing time with the ROSAT satellite observatory has already been granted to further study this interesting object.
Moreover, HE 1104-1805 AB will of course continue to be intensively monitored with ground-based telescopes. The possibility to determine independently the distance scale in the universe is surely a most exciting prospect.
 Quasars are Been as star-like objects in the sky, but observations have shown that they are situated at very large distances. Most astronomers are convinced that they are the exceedingly energetic centers of otherwise normal galaxies and are possibly the powerful manifestations of giant black holes.
 The group consists of Dieter Reimers (leader), Lutz Wisotzki, Thorsten Kohler, Sjur Refsdal and Detlev Groote from the University of Hamburg (Germany), Jean Surdej from the Institut d'Astrophysique, Liege (Belgium), Joseph Wampler (ESO) and others.
ESO EPR Dept