CMB and Molecules at High Redshift

It becomes possible now to detect cold molecules at high redshift in the millimeter domain. Since the first discovery in 1992 by Brown and van den Bout of CO lines at z = 2.28 in a gravitationally lensed starburst galaxy, nearly ten objects are now known to possess large quantities of molecular gas beyond z = 1 and up to z ∼ 5, through millimeter and sub-millimeter emission lines. Even more objects have been detected in their continuum dust emission, and a few galaxies through millimeter absorption lines in front of quasars at z ≤ 1. The continuum dust emission is the most easily detected: for a starburst dust at Td ∼ 60 K, the emission peaks around 60 μm, and falls as λ at longer wavelengths. In the mm domain, the emission is then stronger for the more redshifted objects. For the CO lines, the situation is less favorable, and the reported detections are helped by gravitational amplification. The increase of the CMB temperature Tbg with redshift helps the rotational line excitation (especially at high z), but not its detection. Absorption in front of quasars is a more sensitive probe of cold gas at high redshift, able to detect individual clouds of a few solar masses (instead of 10 M⊙ for emission). Taking advantage of the small size of the QSO, very high spatial resolution (of the order of milli-arcsec) can be achieved, and high spectral (30m/s) resolution, due to the heterodyne technique. The sampled column-densities range between N(H2)= 10 et 10 cm. The high sensitivity allows to detect a multitude of molecular lines in a single object (HCO, HNC, HCN, N2H , CO, CS, H2CO, CN, CCH, H2S etc....), and compare the chemistry with the local one, at z = 0. From the diffuse components, one can measure the cosmic black body temperature as a function of redshift. The high column densities component allow to observe important molecules not observable from the ground, like O2, H2O and LiH for example. All these preliminary studies carry a great hope for what will be observed with future millimeter instruments, and some perspectives are given. I MILLIMETER CO EMISSION LINES AT HIGH REDSHIFT This is a rapidly evolving domain, and at present, only 8 systems are published (cf Table 1). The search of CO lines at high z has been triggered by the detection of the CO(3-2) line in emission in the Faint IRAS source F10214+4724 at z = 2.28 by Brown & Vanden Bout (1991, 1992). At this time, it was a redshift 30 times larger than that of the most distant CO emission discovered in a galaxy. The H2 mass derived was reaching 10h M⊙, with the standard CO-H2 conversion ratio, a huge mass although the FIR to CO luminosities was still compatible with that of other more nearby starbursts. Since then, the derived H2 mass has been reduced by large factors, both with better data and realizing that the source is amplified through a gravitational lens by a large factor (Solomon et al 1992, 1997). After the first discovery, many searches for other candidates took place, but they were harder than expected, and only a few, often gravitationally amplified, objects have been detected: the lensed Cloverleaf quasar H 1413+117 at z = 2.558 (Barvainis et al. 1994), the lensed radiogalaxy MG0414+0534 at z = 2.639 (Barvainis et al. 1998), the possibly magnified object BR1202-0725 at z = 4.69 (Ohta et al. 1996, Omont et al. 1996), the amplified submillimeter-selected hyperluminous galaxy SMM02399-0136 (Frayer et al. 1998), at z = 2.808, and the magnified BAL quasar APM08279+5255, at z = 3.911, where the gas temperature derived from the CO lines is ∼ 200K, maybe excited by the quasar (Downes et al. 1998). Recently Scoville et al. (1997) reported the detection of the first non-lensed object at z = 2.394, the weak radio galaxy 53W002, and Guilloteau et al. (1997) the radio-quiet quasar BRI 1335-0417, at z = 4.407, which has no direct indication of lensing. If the non-amplification is confirmed, these objects would contain the largest molecular contents known (8-10 10 M⊙ with a standard CO/H2 conversion ratio, and even more if the metallicity is low). The derived molecular masses are so high that H2 would constitute between 30 to 80% of the total dynamical mass (according to the unknown inclination), if the standard CO/H2 conversion ratio was adopted. The application of this conversion ratio is however doubtful, and it is possible that the involved H2 masses are 3-4 times lower (Solomon et al. 1997). TABLE 1. CO data for high redshift objects Source z CO S ∆V MH2 Ref line mJy km/s 10 M⊙ F10214+4724 2.285 3-2 18 230 2 1 53W002 2.394 3-2 3 540 7 2 H 1413+117 2.558 3-2 23 330 6 3 MG 0414+0534 2.639 3-2 4 580 5 4 SMM 02399-0136 2.808 3-2 4 710 8 5 APM 08279+5255 3.911 4-3 6 400 0.3 6 BR 1335-0414 4.407 5-4 7 420 10 7 BR 1202-0725 4.690 5-4 8 320 10 8 ∗ corrected for magnification, when estimated Masses have been rescaled to H0 = 75km/s/Mpc. When multiple images are resolved, the flux corresponds to their sum (1) Solomon et al. (1992), Downes et al (1995); (2) Scoville et al. (1997); (3) Barvainis et al (1994); (4) Barvainis et al. (1998); (5) Frayer et al. (1998); (6) Downes et al. (1998); (7) Guilloteau et al. (1997); (8) Omont et al. (1996) It is surprising that very few starburst galaxies have been detected in the CO lines at intermediate redshifts (between 0.3 and 2), although many have been observed (e.g Yun & Scoville 1998, Lo et al 1999). A possible explanation is the lower probability of magnification by lenses in this range (cf Figure 1). II DUST EMISSION IN STAR-FORMING GALAXIES Most of the previous sources, detected in the CO lines, had previously been detected in the dust continuum. At high redshift, it becomes easier to detect the dust emission, because of the large K-correction (e.g. Blain & Longair 1993): the emission is roughly varying as ν with the frequency ν in the millimeter range, until the maximum around 60μm. At one mm, it is even easier to detect objects at z = 5 than z = 1. This has motivated deep searches in blank fields with sensitive instruments, since they should be dominated by high redshift objects, if they exist in sufficient numbers. Their detection will give information about the star-formation rate as a function of redshift, a debated question: the maximum of star-formation rate, found around z =2 from optical studies (Madau et al 1996) could shift to higher z if dust is obscuring the higher-redshift objects. From recent infrared lines observations (Pettini et al 1998), it does not seem a serious problem, however. The first search was made with the SCUBA bolometer on JCMT (Hawaii) towards a cluster of galaxies, thought to serve as a gravitational lens for high-z galaxies behind (Smail et al 1997). The amplification is in average a factor 2. A large number of sources were found, all at large redshifts (z > 1), extrapolated to 2000 sources per square degree (above 4mJy), revealing a large positive evolution with redshift, i.e. an increase of starbursting galaxies. Searches toward the Hubble Deep Field-North (Hughes et al 1998), and towards the Lockman hole and SSA13 (Barger et al 1998), have also found a few sources, allowing to derive a similar density of sources: 800 per square degree, above 3 mJy at 850 μm. This already can account for 50% of the cosmic infra-red background (CIRB), that has been estimated by Puget et al (1996) and Hauser et al (1998) from COBE data. The photometric redshifts of these sources range between 1 and 3. Their identification with optical objects might be uncertain (Richards 1998). However, Hughes et al (1998) claim that the star formation rate derived from the far-infrared might be in some cases 10 times higher than derived from the optical, due to the high extinction. If only some of the sources have a redshift higher than 4, it will flatten the Madau curve at high z. Eales et al (1999) surveyed some of the CFRS fields at 850μm with SCUBA and found also that the sources can account for a significant fraction of the CIRB background (∼ 30%). Their interpretation in terms of the FIGURE 1. H2 masses for the CO-detected objects at high redshift (filled stars), compared to the ultra-luminous-IR sample of Solomon et al (1997, open pentagons), and to the Coma supercluster sample from Casoli et al (1996, filled triangles). There is no detected object between 0.3 and 2.2 in redshift, except the quasar 3c48, marked as a filled dot (Scoville et al 1993, Wink et al 1997). The curve indicates the 3σ detection limit of I(CO) = 1 K km/s at the IRAM-30m telescope (equivalent to an rms of 1mK, with an assumed ∆V = 300km/s). The points at high z can be detected well below this limit, since they are gravitationally amplified. star formation history is however slightly different, in that they do not exclude that the submm luminosity density could evolve in the same way as the UV one. Deep galaxy surveys at 7 and 15μm with ISOCAM also see an evolution with redshift of star-forming galaxies: heavily extincted starbursts represent less than 1% of all galaxies, but 18% of the star formation rate out to z = 1 (Flores et al 1999). III MOLECULES IN ABSORPTION Absorption techniques are also very efficient in the millimeter range, and a few systems have been discovered at high redshift, between z = 0.2 to 1 in the last years (Wiklind & Combes 1994, 95, 96; Combes & Wiklind 1996). The sensitivity is such that a molecular cloud on the line of sight of only a few solar masses is enough to detect a signal, while in emission, upper limits at the same distance are of the order of 10 M⊙. Some general properties of the known absorbing systems are summarised in Table 2. They reveal to be the continuation at high column densities (10–10 cm) of the whole spectrum of absorption systems, from the Lyα forest (10–10 cm) to the damped Lyα and HI 21cm absorptions (10–10 cm). About 15 different molecules have been detected in absorption at high redshifts, in a total of 30 different transitions. This allows a detailed chemical study and comparison with local clouds (Wiklind & Combes 1997, Carilli et al 1998). Up to now, no significant variations in abundances have been found as a function of redshift, at least within the large intrinsic dispersion already existing within a given galaxy. Note that the high redshift allows us to detect some new molecular lines, never observed from the ground at z = 0, such as O2 (Combes et al 1997), H2O or LiH (Combes & Wiklind 1997, 1998). Molecular oxygen has not yet been detected in space, and water vapour appears to be extended and cold (see also the SWAS satellite preliminary results, Melnick et al 1999).