2007ApJ...656..792P

It has recently been argued that the HCN J=1-0 line emission may not be an unbiased tracer of dense molecular gas (n≳10⁴/cm3) in luminous infrared galaxies (LIRGs; L_FIR>10¹¹ L_☉) and that HCO⁺ J=1-0 may constitute a better tracer instead, casting doubt onto earlier claims supporting the former as a good tracer of such gas. In this paper new sensitive HCN J=4-3 observations of four such galaxies are presented, revealing a surprisingly wide excitation range for their dense gas phase that may render the J=1-0 transition from either species a poor proxy of its mass. Moreover, the well-known sensitivity of the HCO⁺ abundance to the ionization degree of molecular gas (an important issue omitted from the ongoing discussion about the relative merits of HCN and HCO⁺ as dense gas tracers) may severely reduce the HCO⁺ abundance in the star-forming and highly turbulent molecular gas found in LIRGs, while HCN remains abundant. This may result in the decreasing HCO⁺/HCN J=1-0 line ratios with increasing IR luminosity found in LIRGs, and it casts doubts on HCO⁺ rather than HCN as a good dense molecular gas tracer. Multitransition observations of both molecules are needed to identify the best such tracer and its relation to ongoing star formation, and to constrain what may be a considerable range of dense gas properties in such galaxies.