Interstellar medium in lensed star-forming galaxies at z ∼2.5
A significant fraction of the star formation density between z = 1–3 has been traced to luminous and ultra-luminous infrared galaxies. Among the most extreme objects seen are those identified via surveys at submillimetre wavelengths; the high infrared luminosities (LIR = 10¹¹−¹³ L⊙) of these “submillimetre galaxies” (SMGs) arise due to the reprocessing of UV and optical light from massive, young stars by interstellar dust, and imply star formation rates SFR= 50–2000M⊙ yr−¹. Such SFRs, combined with the observed increase in number density of SMGs by a factor ∼ 20× out to z = 2 make them candidates to be the progenitors of the most massive “red and dead” elliptical galaxies which dominate the cores of clusters in the present day, yet limitations in the technical capabilities of radio and infrared telescopes have long hindered a detailed understanding of these galaxies, and in particular, the physics of the interstellar medium (ISM). Hence the following key questions remain unanswered: (i) how much molecular gas (H2) do SMGs contain, and what are the likely upper-limits on how long they can sustain their present level of activity?; (ii) do the prodigious luminosities of SMGs stem from a common origin, or does the selection criterion return a “mixed bag” of galaxies?; (iii) what are internal kinematics of SMGs, and to what extent do these influence the global and local star-formation (Schmidt-Kennicutt) law?; (iv) is star-formation in distant SMGs distributed across the entire galaxy, or does it occur in isolated clumps?; (v) what are the typical densities and temperatures of star-forming regions in SMGs, and do they adhere to the observed correlation between far-infrared and radio emission in star-forming galaxies? Recent upgrades to the Karl G. Jansky Very Large Array (VLA), the inauguration of the Atacama Large Millimeter Array (ALMA), and the superlative legacy of the Herschel Space Observatory have jointly provided the toolkit to study the gas and dust emission in distant galaxies in unprecedented detail, and thus to begin to address these fundamental questions. In this thesis, I present ∼120 hours of new VLA observations of ¹²CO J =1−0 emission – the best tracer of molecular gas – in a sample of four lensed SMGs. The combination of high angular resolution with the VLA and the magnifying effects of gravitational lensing allow the ISM properties of these z ∼ 2.5 star-forming galaxies to be seen for the first time on sub-galactic scales, implying gas masses of 10⁹−¹ºM⊙ (subject to the 12CO-luminosity-to-H2-mass conversion factor, αCO), and demonstrating the presence of an extended, low-excitation gas reservoir. In conjunction with observations of the excited gas phase from the Plateau de Bure Interferometer (PdBI), these new data point to variations in the densities and temperatures of H2 throughout each galaxy. The wide bandwidth and phenomenal sensitivity of the VLA yields the first detections of 115-GHz continuum emission at high redshift, which I use in conjunction with well-sampled dust spectra from Herschel and data at longer wavelengths from the VLA to decompose the radio spectra of two galaxies into contributions arising from thermal dust emission, optically-thin free-free emission from Hii regions, and non-thermal synchrotron emission. From these measurements, estimates of SFRradio are made, providing an independent check on SFRIR, and enabling the degeneracy between the heating of dust due to star formation and that due to hidden AGN activity to be broken. Via this spectral index de-convolution, I find L[AGN]/LIR fractions of 35% and 55% for the two SMGs, in broad agreement with previously published estimates for these sources based on their mid-IR spectral properties. In the exceptional case of SMMJ23152–0102, magnification by the foreground cluster is so extreme (∼ 32.5×) that the VLA synthesised beam traces regions of order ∼ 130 pc in the source plane, and identifies a series of cool, dense clumps (Tk = 30–70 k; log[n(H2)/cm−3]= 3.6–3.9) within the gas reservoir, which contain between them 10–60% of the total molecular gas of the system. These clumps are offset from the far-infrared/radio correlation, which I argue has implications for their ages.