¹ Imperial College, University of London, U.K.
² University of Cambridge, U.K.
³University of Hertfordshire, U.K.,
⁴Gemini North, Hawaii, U.S.A.
⁵Instituto de Astrofisica de Canarias, Spain
⁶University of Leicester, U.K.
⁷University of Exeter, U.K.
⁸University of Hawaii, U.S.A.

One of the main factors that influenced the design of the LAS was the opportunity to search for rare objects, extending the work of 2MASS in finding very cool brown dwarfs, and of SDSS in finding quasars of very high redshifts, as well as cool brown dwarfs. These goals are described in Lawrence et al. (2006), and Hewett et al. (2006). UKIDSS DR1, which occurred at the end of July, provides the first opportunity for teams to exploit a dataset sufficiently large to be of interest. The LAS coverage in DR1 is 190 sq degs.

The coolest brown dwarfs are the T dwarfs, of which 99 are known, all discovered since 1995. The main samples have come from SDSS, where candidates are identified as extremely red in i-z, or 2MASS, where, paradoxically they are selected as very blue in J-K, or J-H. Two preliminary spectral classification schemes (one from each survey) were published, and these were recently merged. This revised classification scheme (Burgasser et al., 2006) defined nine spectral classes from T0 to T8. The primary spectral standard for the coolest class, T8, is the object 2MASS 0415-09. There are only six T8 dwarfs known. These are the coolest brown dwarfs and have temperatures ~700 K.

Figure 1: The first T dwarf from UKIDSS followed up at UKIRT. The classification T4 follows the scheme of Burgasser et al. (2006).

Jupiter has a temperature ~150 K. What lies in between, in the temperature range T~ 150-700 K? One of the goals of UKIDSS is to explore this temperature range. Models provide a guide to how to discover these ultra-cool dim dwarfs, but there are significant uncertainties. Ultra-cool dwarfs are expected to be extremely red in z-J, and so difficult to detect in z. They may also continue to get bluer in J-H, and therefore be faint in H. Therefore the new Y filter (0.97-1.07 micron), between z and J, is expected to play an important role in this work. At some point a new spectral feature is expected to emerge, possible NH3 absorption, defining a new spectral class. Thanks to the foresight of Davy Kirkpatrick, the class is already named: they will be called Y dwarfs.

Roughly speaking DR1 surveys about 1/3 of the volume of 2MASS (depending on how the search is done, i.e. which filter limits the depth), so DR1 should contain a number of cool T dwarfs. Bearing in mind the uncertainty in the predictions a range of strategies has been followed, and is already bearing fruit. So far we have confirmed four T dwarfs spectroscopically, at UKIRT, Subaru and Gemini, with classifications T4, T4, T6 and T8, taking the T dwarf count to over 100. Fig. 1 shows the spectrum of the T4 confirmed at UKIRT, with UIST. The discovery of a T8 dwarf, as cool as any brown dwarf known, so early on in the follow-up of DR1 sources is extremely encouraging. Some additional interesting candidates around 9-10h RA are just becoming observable as this goes to press...

Figure 2: The discovery spectrum of the first very high redshift quasar discovered in UKIDSS (from Venemans et al, in prep). This 1200s spectrum was taken on the night of 1st Sept, 2006, with FORS2 on the VLT.

SDSS has been highly successful in discovering a number of quasars beyond z=6. The most distant quasar at z=6.4, found by SDSS, lies near the observable limit of the survey. Due to absorption by intervening neutral hydrogen, at higher redshifts a quasar would be extremely faint in z, the longest-wavelength SDSS band. This has brought about an impasse in the search for quasars of higher redshift, and it is notable that the z=6.4 quasar was discovered four years ago now. Yet analysis of the very strong absorption in the Lya forest of the highest redshift quasars has yielded tantalising evidence that at z=6 we have reached the tail-end of the epoch when the Universe was reionised. Therefore there is strong motivation for extending the redshift limit of quasar surveys. By finding quasars beyond z=6.4 it will be possible to explore the conditions in the intergalactic medium and thereby chart the progress of reionisation. At present there is no consensus on when this period was, or how long it lasted, or what type of source was responsible for reionisation.

The search for high-redshift quasars will also benefit from the Y band. Indeed the Y band wavelength range was carefully designed with this goal in mind. Quasars at z>6.4 will be very red in i-Y or z-Y, but bluer in Y-J than the more common L and T brown dwarfs, and therefore distinguishable from them. To a limit Y=19.5 we expect to find about one quasar z>6.0 in 150 sq degs, so DR1 is promising for this search. Our results so far are consistent with this expectation. We have searched the majority of DR1, and have found a single high- redshift quasar, at z=5.86. The spectrum is shown in Fig. 2, and shows the characteristic very strong break in the continuum across Lyα. The source is detected at S/N>10 in the Y band, but is undetected in SDSS, demonstrating the power of the UKIDSS LAS. As with the brown dwarfs, the discovery of this high-redshift quasar is extremely encouraging for the future of the search, as the LAS database expands.

References:
Burgasser A. et al., 2006, ApJ 637, 1067
Hewett P. et al., 2006, MNRAS 367, 454
Lawrence A. et al., 2006, MNRAS submitted (astro-ph/0604426).

¹Instituto de Astrofisica de Canarias, Tenerife, Spain,
²Univ. of Southampton, U.K.,
³Univ. of Sheffield, U.K.,
⁴ESO, Chile

In the past decade or so overwhelming evidence has pointed to a clear coupling between accretion and the formation of relativistic jets in galactic X-ray binary systems (see Fender 2006). Accretion states associated with hard X-ray spectra appear to be associated with the production of a relatively steady, continuously replenished and partially self-absorbed outflow, while major outbursts are associated with more discrete ejection events which may be resolved and tracked with radio interferometers (e.g. Mirabel & Rodriguez 1994). Furthermore the past few years have seen the first quantitative scalings for jet production and power between the black holes in X-ray binary systems and the supermassive black holes in active galactic nuclei (e.g. Merloni et al. 2003).

In the radio band the steady jets observed during hard X-ray states have a flat (α ~ 0, where Sν ∝ &nu^α ) spectrum, probably resulting from self-absorption in a self-similar outflow. Above some frequency this flat spectral component should break to an optically thin spectrum (α ~ -0.6) corresponding to the point at which the entire jet becomes transparent; i.e. emission at the break frequency arises primarily from the `base' of the jet. There is some evidence from some black hole X-ray binaries that this break occurs around the near infrared spectral region (e.g. Corbel & Fender 2002), something which can be well fit by jet models. Furthermore, for a few X-ray binaries (GRS1915+105 and 4U0614+091), there is essentially no doubt that infrared synchrotron emission has been observed. Thus the case is strong that there is a significant contribution of synchrotron emission, probably optically thin, in the near-infrared spectral regimes of X-ray binaries. However, one key test which is yet to be reported is a measurement of the linear polarization in this regime. Not only would a high level of linear polarization confirm the synchrotron interpretation, it would offer us the opportunity to study the degree of ordering and orientation of the magnetic field at the base of the jet.

In 2004 we obtained HK (1.4 μm to 2.5 μm) linear spectro-polarimetry of three X-ray binaries, Sco X-1, Cyg X-2 and GRS1915+105, using UKIRT and UIST+IRPOL2. Figure 1 shows our near-infrared linear polarization values. For Sco X-1 the mean linear polarization is 0.30(0.04)% and 0.64(0.05)% at 1.65 μm and 2.4 μm respectively. Although the H-band polarization may be described an interstellar, the K-band infrared polarization clearly cannot. Similarly, for Cyg X-2, our IR linear polarization measurements of 1.7% and 6.1% at 1.65 μm and 2.4 μm respectively show a considerable excess in the near-infrared compared to the optical. It is clear that interstellar polarization cannot explain the observed optical and near-infrared polarization values. Cyg X-2, like Sco X-1, is a radio source and member of the `Z-source' and as such is also very likely to be a jet source (although one has never been spatially resolved).

Figure 1: From top to bottom; UIST+IRPOL2 HK linear polarization spectrum of GRS1915+05, Cyg X-2, Sco X-1 and the polarised standard star HD184143.	Figure 2: Expectations for the linear polarization signature in the near-infrared and optical regimes of X-ray binaries. The key spectral points are the break from optically thick (low polarization ~10%) to optically thin (high polarization) synchrotron emission, and the point at which the thermal (low polarization) emission begins to dominate over the jet.

An attractive possibility is that the `excess' near-infrared polarization is due to synchrotron emission from jets. Two polarization signatures are expected from the jet, depending on whether or not the synchrotron emission is optically thick or thin. Above some frequency this flat spectral component should break to an optically thin spectrum corresponding to the point at which the entire jet becomes transparent. For the optically thick part of the spectrum, no more than a few percent polarization is expected. The most exciting prospect is that of a large fractional polarization from optically thin synchrotron emission. While it is by no means firmly established, a small number of observational and theoretical results suggest this change in polarization should occur around the near-infrared band. Optically thick synchrotron emission has a maximum linear polarization of ~10%. Optically thin synchrotron emission can have a fractional linear polarization as high as 70%. Nevertheless, some X-ray binary jets show up to ~30% polarization in the radio, indicating a highly significant ordering of the magnetic field.

Figure 2 illustrates in more detail our expectation for the intrinsic (i.e. before interstellar scattering) linear polarization signature in the near-infrared and optical regimes, based on published spectral energy distributions and the simple ideas outlined above about the expected polarization fraction. At long wavelengths (maybe in the mid-infrared: the break is hard to determine precisely) there will only be ~1% polarization from the self-absorbed jet; at short wavelengths a comparable level will be measured due to scattering in the accretion flow. However, in the relatively narrow spectral region in which optically thin synchrotron emission dominates, we may expect a strong signature which initially rises to longer wavelengths as the relative jet:disc fraction increases. The measurements reported here, which indicate a significant contribution of synchrotron emission in the near infrared, should be the beginning of a highly fruitful line of enquiry in the near future.

References:
Corbel S., Fender R.P., 2002, ApJ, 573, L35
Fender R.P., 2006, in Compact Stellar X-ray Sources, eds. W. Lewin and M. van der Klis, Cambridge University Press, Cambridge, (astro-ph/0303339)
Mirabel I.F., Rodriguez L.F., 1994, 371, 46
Merloni A., Heinz S., di Matteo T., 2003, MNRAS, 345, 1057

¹ University of Arizona, U.S.A.
² Royal Observatory of Belgium

Rho Casseopeiae is a naked-eye pulsating yellow hypergiant, one of the most luminous stars in the Galaxy. Cool hypergiants are very rare 20-50 solar mass stars, and considered to be the late-type sisters of the Luminous Blue Variables. They exist near a dynamically unstable region in the upper HR diagram called the "Yellow Evolutionary Void". They evolve bluewards and are possible progenitors of core-collapse Supernovae. Once about every 50 years Rho Cas undergoes an outburst when it dims by more than a full visual magnitude while the entire surface cools by ~4000 K. This behavior is reminiscent of the brightness declines in R CrB stars that are less massive, hydrogen-poor Asymptotic Giant Branch stars. The cause of the enigmatic stellar outbursts remains a long-standing puzzle in astrophysics.

Figure 1. Bottom panel: visual brightness curve of Rho Cas from AAVSO showing the great outburst of 2000-01. Top panel: UKIRT CGS4 spectrum observed at outburst (red color) reveals the CO band in absorption with strong atomic lines typical of a K-type star (marked in black). Shown for comparison is the IRTF spectrum observed around maximum brightness (blue color) when the star was in a variability phase with strong emission lines.

While most studies in the past of the cool hypergiant have concentrated on the optical wavelength region, we searched for clues for the erratic behavior of Rho Cas in the near-IR. We retrieved two spectra from the UKIRT Archive observed with the CGS4 spectrograph with a spectral resolution of 1000-3000. The first spectrum was observed in October 1998. Apparently it was observed for telluric correction purposes only, because it had been archived under the BS 9045 name of the hypergaint. To our amazement, we detected very strong emission in the first overtone band of CO. Lobel et al. (2003; ApJ 583, 923) observed strong emission in optical atomic lines during the same month. We carefully monitored the CO band with other telescopes which confirmed that the prominent emission lines appear at the beginning of each pulsation cycle when the atmosphere rapidly expands, or just before the optical brightness begins to decrease. The second UKIRT spectrum was observed in the fall of 2000 when the hypergiant went in strong outburst (Figure 1). This spectrum is the first IR spectrum ever observed of a luminous cool hypergiant in outburst. It confirmed independently that the effective temperature of the mysterious hypergiant dramatically decreased from F to the late-K spectral type in about 200 days, exhibiting strong low-excitation lines of Ti that originate in circumstellar gas expelled during the outburst and observed in blue-shifted absorption bands of TiO with optical spectra.

The strong molecular CO band is very peculiar for an F-type star with relatively large Teff ~ 7000 K. Observations of the hypergiant in the 1980-ies did not provide optical spectra of sufficiently large spectral resolution to link the cyclic emission we find in the CO bands with variable emission observed in optical atomic lines. These older studies only proposed that the optical emission lines would perhaps originate from a stellar chromosphere, while the CO lines would be formed in very extended circumstellar gas shells. These studies remained inconclusive because classical chromospheric indicators such as emission in Ca II or Mg II resonance lines could never be observed in the cool hypergiant, and remote circumstellar gas shells remain unobserved with direct imaging.

A good opportunity to re-investigate the issue firsthand came when our optical spectral monitoring program signaled the onset of a new line emission episode in July 2004. We contacted the UKIRT service observing team (C. Davis and S. Leggett) and asked them to observe an echelle spectrum to check if the IR CO band appeared in emission as well. And sure enough, it did! Not only did we observe strong CO emission lines, they were also overlaid on top of broader absorption lines, and the detailed line shapes closely matched the profiles of optical atomic lines in spectra observed only a month later.

Figure 2. Diagram of the circumstellar environment of Rho Cas based on optical and IR spectroscopic features and weak IR dust emission observed with IRAS.

Where does the molecular emission originate from in the yellow hypergiant? The optical atomic emission lines have earlier been proposed to originate from an interface between the fast and collimated stellar wind colliding with material previously expelled during the hypergaint's violent mass-loss history. After analyzing the new spectra, we published a paper (astro-ph/0607158) that proposes a consistent model for the CO emission formation region as well. The variable molecular emission emerges from the cooling flow behind a pulsation-driven shock wave at a relatively small distance from the stellar photosphere (Fig. 2). Interestingly, the eruptive R CrB stars also show two groups of emission lines that behave spectacularly similar to those observed in erratic Rho Cas. Can the F-type hypergiant Rho Cas blow dusty clouds similar to its smaller siblings? The upcoming observations with the Spitzer Observatory promise exciting new insights and answers to many of these questions.

¹ AIU Jena, Germany
² ESO, Garching, Germany
³ Tel Aviv University, Israel

Figure 1. UFTI image of HD3651B (circled star), the first directly imaged brown dwarf companion of the exoplanet host star, HD3651.

In the course of near infrared imaging programs carried out at UKIRT and at the La Silla observatory during recent years, several new companions of exoplanet host stars have been detected. These observations help to improve our knowledge of the impact of stellar multiplicity on the planet formation process.

A new faint companion of the exoplanet host star HD3651 has recently been detected, located 43 arcsec north-west of the star at a physical projected separation of 480 AU. This companion, HD3651B, was first imaged with UKIRT and its infrared camera UFTI. With follow-up observations, carried out at UKIRT and the La Silla observatory, we have proven that the companion shares the proper motion of the exoplanet host star.

HD3651B is a faint source in the H-band, but is even fainter in the visible spectral range, which is typical for the spectral energy distribution of brown dwarfs. According to evolutionary models of substellar objects, the infrared photometry of HD3651B is consistent with a brown dwarf with a mass between 20 and 60 M_Jup (age dependent) and a temperature between 800 and 900 K. Such cool substellar objects are also referred to as T dwarfs, showing strong methane absorption bands in their spectra.

HD3651B is the first directly imaged brown dwarf which revolves around an exoplanet host star. The HD3651 system is an interesting example that might prove that planet and brown dwarf formation can occur around the same star. The star and its brown dwarf could emerge from a fragmentation process of a large gas and dust cloud, while the planet probably forms in a disc around the star. HD3651B is the faintest directly imaged companion of an exoplanet host, as well as one of the faintest brown dwarfs yet detected in the solar vicinity. Today there are only two further brown dwarfs known with a comparable brightness, namely Gl570D and 2MASS J0415-0935.

References:
M. Mugrauer, A. Seifahrt, R. Neuhaeuser, T. Mazeh, 2006 MNRAS Letters, astro-ph/0608484