Max Planck Institute for Astronomy - Annual Report 2007

26 II. Highlights
F 1 [5]
–2
0
2
4
6
8
10
12
14
0
a
HD172555A H = 4.3
Eri 2004oct H = 1.9
Ind A 2004oct H = 2.3
HD114613 H = 3.3
HD17925 H = 4.2
0.5 1
Distance from primary
1.5 2
Fig. II.2.4: Brightness contrast achieved depending on distance
to star. a) Stars with a brightness of H 4.5, b) Stars with
7.5 H 6.5.
This would not have been possible with a conventional
exposure with adaptive optics (right).
In the end, the astronomers were not able to detect
any companions to the 54 stars. However, even with this
negative result, new insights can be extracted if one fully
understands the sensitivity of the instrument. For this a
strict analysis is required.
For one thing, the brightness contrast achieved in
the images between star and planet was established in
dependence on the mutual distance. This contrast is not
dependent on the star’s apparent brightness but is dependent
on “Seeing” and thus requires an individual analysis
of all images. These values were then compared to current
models for exoplanet spectra. According to current
theories, the temperature of a giant planet which is older
than ten million years should have sunk to 800 K and it
should possess a T 8 or later spectral type.
Fig. II.2.4 a is an example of stars which are brighter
than 4.5 mag in the H-Band with the achieved brightness
contrast having a significance of 5 s. Accordingly, in a
distance range of one half to one arcsecond, a contrast of
10 to 12 mag was achieved. This corresponds to intensity
ratios of between 10,000 and 60,000 and thus surpasses
the above-mentioned superspeckle-interference limit by
more than one order of magnitude. These are currently
the most contrast rich images of methane-rich companions
that have ever been taken from space or earth. Stated
differently, a brightness contrast of 10 mag at a distance
of 0.5 arc seconds could be achieved for 45 percent of
the stars in the survey. Fig. II.2.4b highlights the loss of
contrast for stars with 7.5 H 6.5.
These technical characteristics could then be compared
with the mentioned models for planet spectra in
order to make statements regarding the detection limits
for these companions. Said more simply, the astronomers
should have been able to discover a gas planet of at least
F 1 [5]
–2
0
2
4
6
8
10
12
14
0
b
AO Men H = 7.0
ERX HD 8558 H = 6.9
Hip 112312A H = 7.2
Hip 30043 AB pic H = 7.1
KW Lup H = 6.6
BD05 378 H = 7.2
GJ 207.1 H = 7.1
ERX8 HD 9054 H = 6.9
Hip 9141 H = 6.6
0.5 1
Distance from primary
1.5 2
five Jupiter masses at a distance of 24 AU from the star
and a planet with at least ten Jupiter masses more than
9 AU from the star (Fig. II.2.5) – had there been one.
However, the distance of a planet projected on the celestial
plane to the central star varies with its position on
the orbit. Only twice per orbit does it reach the greatest
angular distance from the star. In order to take this effect
into account, the astronomers estimated the orbital movement
of 10,000 hypothetical planets for each observed
star whereby they varied the mass, the large semiaxis
of its orbit, and its eccentricity within specific limits. In
combination with the contrast curves (Fig. II.2.4), the
probability of a possible discovery was determined for
each star.
Fig. II.2.5: Minimal distance at which among ten selected stars
a planet with five Jupiter masses would still be detectable.
AO Men
CD –64 1208
GJ 182
GJ 799A
GJ 799B
GJ 803
HD 155555AB
LQ Hya
Hip 23309
HD 155555C
LQ Hya, Age = 13 Myr
GJ 182, Age = 12 Myr
Burrows et al. 2003
Baraffe et al. 2003
Masciadri et al. 2005
0 10 20 40 60
Projected Separation [AU]