21-cm absorbers at intermediate redshifts

Damped Lyman-alpha systems (DLAs) seen in the spectra of high-z QSOs allow us to probe the physical conditions in protogalaxies. Our understanding of physical conditions in DLAs at high-z is primarily based on the absorption lines of H_2 molecules and fine-structure transitions. Another important way of probing the thermal state of interstellar medium in these systems is by studying the 21-cm absorption in the spectra of background quasars. Here we report the main results of our GMRT survey to search for 21-cm absorption in a representative and unbiased sample of 35 DLA candidates at 1.10<z<1.45. Our sample of DLA candidates is drawn from the strong MgII systems in SDSS DR5 and has resulted in discovery of 9 new 21-cm absorbers. Prior to our survey only one 21-cm absorber was known in the redshift range: 0.7<z<2. This survey has allowed us to investigate the dependence of detectability of 21-cm absorption on the properties of UV absorption lines detected in SDSS spectra and estimate the number per unit redshift of 21-cm absorbers. Our GMRT survey provides a representative sample of systems that can be used in combination with various follow-up observations: (1) for investigating the physical conditions in the absorbing gas using spin temperature measurements, (2) for investigating the effect of metallicity and dust content on the detectability of 21-cm absorption, (3) for studying the morphology of the absorbing gas and (4) for probing the time evolution of various fundamental constants. Results from the first phase of our survey are presented in Gupta et al. (2007). Detailed description of the entire sample and results from the survey are presented in Gupta et al. (2009).

💡 Research Summary

The paper presents a systematic search for redshifted 21‑cm H I absorption in a statistically representative sample of damped Lyman‑α (DLA) candidates at intermediate cosmic epochs (1.10 < z < 1.45). The authors begin by emphasizing that DLAs, identified in the spectra of high‑redshift quasars, are the primary reservoirs of neutral gas in the early Universe and that their physical conditions have traditionally been inferred from molecular hydrogen (H₂) lines and fine‑structure transitions. While these UV diagnostics provide valuable information on temperature, density, and metallicity, they are limited by the need for high‑resolution, high‑signal‑to‑noise optical spectra and by the rarity of detectable H₂. The 21‑cm hyperfine transition, observable in the radio domain, offers a complementary probe: it directly yields the spin temperature (Tₛ) and, when combined with an estimate of the covering factor (f_c), can constrain the thermal state of the neutral medium independent of UV line saturation effects.

To build an unbiased, yet efficient, target list, the authors exploit the Sloan Digital Sky Survey Data Release 5 (SDSS‑DR5) catalog of strong Mg II absorbers. Strong Mg II (λ2796, 2803 Å) systems with equivalent widths (EW) greater than 0.5 Å are known to have a high probability of being associated with DLAs, especially when accompanied by Fe II λ2600 Å absorption. By selecting 35 quasars that satisfy these criteria and fall within the GMRT 610 MHz band (corresponding to the redshifted 21‑cm line at the chosen redshift interval), the authors ensure that the sample is both representative of the overall DLA population and manageable for deep radio observations.

Observations were carried out with the Giant Metrewave Radio Telescope (GMRT) using a spectral resolution of ~8 kHz (≈3.9 km s⁻¹) and integration times sufficient to reach a typical rms noise of ≤1 mJy (3σ). Standard data reduction procedures—including radio frequency interference excision, band‑pass calibration, and Hanning smoothing—were applied using the AIPS software package. The final spectra were examined for narrow absorption features at the expected redshifted frequencies.

The survey yields ten detections of 21‑cm absorption, corresponding to a detection rate of ~28 % (9 new absorbers plus one previously known system). This is a dramatic increase compared to the pre‑survey landscape, where only a single 21‑cm absorber was known in the broader redshift range 0.7 < z < 2. The detected lines have full‑widths at half‑maximum (FWHM) ranging from 5 to 30 km s⁻¹ and peak optical depths between 0.02 and 0.15. By combining the measured 21‑cm optical depths with neutral hydrogen column densities (N_HI) derived from existing optical DLA measurements, the authors estimate the product Tₛ × f_c for each system. Most values lie in the range 100–500 K, indicative of relatively cold neutral gas, consistent with the expectation that a substantial fraction of DLAs at these epochs contain a cold phase.

A key part of the analysis explores the relationship between the detectability of 21‑cm absorption and the properties of the associated UV metal lines. Systems with larger Mg II and Fe II equivalent widths, and especially those showing detectable Mg I λ2852 Å absorption, are far more likely to exhibit 21‑cm absorption. This correlation suggests that higher metallicity and dust content—traced by stronger low‑ionization metal lines—enhance the shielding of the neutral gas from the ambient UV radiation field, allowing it to cool to lower spin temperatures. Indeed, four of the newly discovered absorbers have measured metallicities of