Outflows of hot molecular gas in ultra-luminous infra-red galaxies mapped with VLT-SINFONI

TL;DR

This study maps hot molecular gas outflows in nearby ULIRGs using VLT-SINFONI, revealing their properties, velocities, and implications for galaxy evolution and star formation, with most gas remaining bound within the systems.

Contribution

First detailed morphological characterization of hot molecular outflows in ULIRGs using near-IR integral-field spectroscopy.

Findings

Outflows detected in 3 of 4 ULIRGs analyzed.

Outflow velocities range from 300 to 900 km/s.

Most outflowing gas remains gravitationally bound within the galaxy.

Abstract

We present the detection and morphological characterization of hot molecular gas outflows in nearby ultra-luminous infrared galaxies, using the near-IR integral-field spectrograph SINFONI on the VLT. We detect outflows observed in the 2.12 micron H$_{2}$ 1-0 S(1) line for three out of four ULIRGs analyzed; IRAS 12112+0305, 14348-1447, and 22491-1808. The outflows are mapped on scales of 0.7-1.6 kpc, show typical outflow velocities of 300-500 km/s, and appear to originate from the nuclear region. The outflows comprise hot molecular gas masses of ~6-8x10$^3$ M(sun). Assuming a hot-to-cold molecular gas mass ratio of 6x10$^{-5}$, as found in nearby luminous IR galaxies, the total (hot+cold) molecular gas mass in these outflows is expected to be ~1x10$^{8}$ M(sun). This translates into molecular mass outflow rates of ~30-85 M(sun)/yr, which is a factor of a few lower than the star formation rate in these ULIRGs. In addition, most of the outflowing molecular gas does not reach the escape velocity of these merger systems, which implies that the bulk of the outflowing molecular gas is re-distributed within the system and thus remains available for future star formation. The fastest H$_{2}$ outflow is seen in the Compton-thick AGN of IRAS 14348-1447, reaching a maximum outflow velocity of ~900 km/s. Another ULIRG, IRAS 17208-0014, shows asymmetric H$_{2}$ line profiles different from the outflows seen in the other three ULIRGs. We discuss several alternative explanations for its line asymmetries, including a very gentle galactic wind, internal gas dynamics, low-velocity gas outside the disk, or two superposed gas disks. We do not detect the hot molecular counterpart to the outflow previously detected in CO(2-1) in IRAS 17208-0014, but we note that our SINFONI data are not sensitive enough to detect this outflow if it has a small hot-to-cold molecular gas mass ratio of < 9x10$^{-6}$.