A 22-GHz search for molecular absorption at z ∼ 3 with the upgraded ATCA

Mon. Not. R. Astron. Soc. 340, 139–142 (2003) A 22-GHz search for molecular absorption at z ∼ 3 with the upgraded ATCA S. J. Curran,1  M. T. Murphy,1 J. K. Webb1 and Y. M. Pihlström2 1 School of Physics, University of New South Wales, Sydney, NSW 2052, Australia 2 National Radio Astronomy Observatory, Socorro, NM 87801, USA Accepted 2002 November 13. Received 2002 October 31; in original form 2002 September 25 ABSTRACT Key words: galaxies: ISM – quasars: absorption lines – early Universe – radio continuum: galaxies. 1 INTRODUCTION Molecular absorption lines at high redshift can provide an excellent probe of cosmological physics such as the cosmic microwave background, values of the fundamental constants and the chemistry of the early Universe (e.g. Wiklind & Combes 1996c, 1997, 2001; Drinkwater et al. 1998). However, such studies are limited to the four known high-redshift molecular absorption systems, towards TXS 0218+357 (Wiklind & Combes 1995), PKS 1413+135 (Wiklind & Combes 1997), TXS 1504+377 (Wiklind & Combes 1996a) and PKS 1830–211 (Wiklind & Combes 1998). In the search for new systems, one systematic approach is to target high column density absorbers with known redshifts. A convenient sample is the damped Lyman-alpha absorbers, which have neutral hydrogen column densities N HI
1020 cm−2 . In order to select a sample, we produced a catalogue of all known DLAs (Curran et al. 2002b)1 and shortlisted those which are illuminated by radio-loud quasars (i.e. those with a measured radio flux density >0.1 Jy). From this sample of 57 we selected those which have 12- or 3-mm fluxes. Recently, we completed a search for molecular absorption towards 11 DLAs at λ  3 mm with the SEST 15-m and Onsala 20-m telescopes which, apart from one tentative detection, only lead to upper limits for 18 transitions (Curran et al. 2002a). With the upgraded ATCA it is now possible to improve on these previous attempts. In this paper we present the results of our first search with this  E-mail: 1 A version of this catalogue is continually updated on-line and is available from http://www.phys.unsw.edu.au/∼sjc/dla. C 2003 RAS telescope – observations towards the known southern centimetreloud quasars occulted by DLAs in which a commonly detected transition falls into the 12-mm band. 2 O B S E RVAT I O N S The observations were performed in 2002 June with the ATCA at Narrabri, Australia during excellent weather conditions which gave good phase stability. The telescope has recently been upgraded with the addition of 12- and 3-mm receivers to antennas 2, 3 and 4 (see Wong & Melatos 2002), thus permitting the search for redshifted molecular rotational lines with this instrument. As mentioned above, the 3-mm band has been used quite extensively in previous searches towards DLAs (Wiklind & Combes 1994b, 1995, 1996b; Curran et al. 2002a), although to date no 12-mm searches have been published. At this wavelength we are able to take advantage of the lower system temperature (≈60 K) and better atmospheric stability to observe towards the southern (δ  30◦ ) quasars of sufficient flux2 occulted by DLAs at redshifts of ∼3 (Table 1).3 In order to minimize the bias introduced by using 2 Originally PKS 0336–017 was confused with Q 0336–019 (J 0339–017), which was then used as a calibrator for the source (Table 1). The measured 22-GHz flux density of S 22 = 0.15 Jy now joins S 0.4 = 1.31, S 1.4 = 0.60, S 2.7 = 0.45 and S 5.0 = 0.30 Jy for the measured radio flux densities of 0336–017 (Curran et al. 2002b). 3 HCO+ is the strongest and most commonly detected molecule in the four known high-redshift molecular absorption systems (see Wiklind & Combes references). Note that no CO transitions fell into either of the two sub-bands at these redshifts. We report a λ
1 cm search for rotational molecular absorption towards quasars, now possible with the upgraded Australia Telescope Compact Array (ATCA). The targets were PKS 0201+113, PKS 0336–017 and Q 0537–286, where known damped Lyman-alpha absorption systems (DLAs) could cause redshifted molecular absorption in the 12-mm band of the telescope. We place 3σ upper limits on any HCO+ 0 → 1 absorption features of <30 mJy per 3.4 km s−1 channel. The non-detections could be attributed to the inherent low metallicities in DLAs leading to generally low H2 , and thus HCO+ , column densities. In general, the detection of molecular rotational transitions in DLAs could be further hindered by a lower than expected CO-to-H2 conversion ratio, whether due either to photoionization of carbon or to its relative under-abundance at high redshift. 140 S. J. Curran et al. Table 1. The 22-GHz illuminated DLAs where the redshifted HCO+ 0 → 1 falls into either the 16.089–18.888 or 20.089–22.488 GHz ATCA bands. z abs is the DLA redshift and ν obs is the expected frequency of the redshifted HCO+ 0 → 1 line, given to the number of significant figures available from the optical data. The red, visible and blue magnitudes of the quasar are given as well as the measured 22-GHz flux densities from both the literature (Lit.) and our observations (Obs.). Any difference between these two values may be due to quasar variability. Source 0201+113 0336–017 0537–286 Coordinates (J2000) hms d 02 03 46.7 03 39 00.9 05 39 54.3 11 34 44 −01 33 18 −28 39 56 z abs 3.38639 3.0619 2.976 ν obs 20.331 21.958 22.43 Magnitude V R ≈S 22 (Jy) Lit.a Obs. Calibrator B 21.7 19.5 19.8 19.5 18.8 19.0 18.8 19.1 18.9 0.55 – 1.66 0202+149 2.00 0336–019 3.37 Self calibration 0.59 0.15 0.58 ≈S 22 (Jy) Lit. Obs. 1.41 2.2 a The flux density for 0201+113 is from Teräsranta et al. (2001) and for 0202+149, 0336–019 and 0537–286 from the ATCA calibrators site (http://www.narrabri.atnf.csiro.au/calibrators/). The uncertainty of ≈15 per cent in these calibrators is the major contributor to the errors in the measured flux densities. Figure 2. HCO+ 0 → 1 at z = 3.062+0.011 −0.007 towards 0336–017. The 1σ rms noise is 6 mJy. The overlap between the two spliced bands is 12 MHz. 3 R E S U LT S A N D D I S C U S S I O N In Figs 1 to 3 we show the time averaged spectra over both good baselines and note that there are no HCO+ 0 → 1 absorption features of 3σ per 3.4 km s−1 channel in these DLAs. From the r.m.s. noise levels we derive optical depth limits for a resolution of 1 km s−1 and
column density limits according Nmm ∝ τ dv for an excitation temperature of ≈10 K (see Curran et al. 2002a). These are listed in Table 2 together with all previously published results. Figure 3. HCO+ 0 → 1 at z = 2.976+0.010 −0.004 towards 0537–286. The 1σ rms noise is 8 mJy. The gap in the spectrum occurs since there was no bandpass calibration and the fitted polynomial is unreliable at these frequencies. From the table we see that, after 3–5 h per source, our limits are among the lowest of all the searches and we achieve the most sensitive search for HCO+ absorption in a DLA published to date. However, there are unce (...truncated)