Spitzer quasar and ulirg evolution study (quest). IV. Comparison of 1 Jy ultraluminous infrared galaxies with palomar-green quasars

We report the results from a comprehensive study of 74 ultraluminous infrared galaxies (ULIRGs) and 34 Palomar-Green (PG) quasars within z ∼ 0.3 observed with the Spitzer Infrared Spectrograph (IRS). The contribution of nuclear activity to the bolometric luminosity in these systems is quantified using six independent methods that span a range in wavelength and give consistent results within ∼ ± 10%-15% on average. This agreement suggests that deeply buried active galactic nuclei (AGNs) invisible to Spitzer IRS but bright in the far-infrared are not common in this sample. The average derived AGN contribution in ULIRGs is ∼35%-40%, ranging from ∼15%-35% among "cool" (f ₂₅/f ₆₀ ≤ 0.2) optically classified H II-like and LINER ULIRGs to ∼50 and ∼75% among warm Seyfert 2 and Seyfert 1 ULIRGs, respectively. This number exceeds ∼80% in PG QSOs. ULIRGs fall in one of three distinct AGN classes: (1) objects with small extinctions and large polycyclic aromatic hydrocarbon (PAH) equivalent widths are highly starburst-dominated; (2) systems with large extinctions and modest PAH equivalent widths have larger AGN contributions, but still tend to be starburst-dominated; and (3) ULIRGs with both small extinctions and small PAH equivalent widths host AGN that are at least as powerful as the starbursts. The AGN contributions in class 2 ULIRGs are more uncertain than in the other objects, and we cannot formally rule out the possibility that these objects represent a physically distinct type of ULIRGs. A morphological trend is seen along the sequence (1)-(2)-(3), in general agreement with the standard ULIRG-QSO evolution scenario and suggestive of a broad peak in extinction during the intermediate stages of merger evolution. However, the scatter in this sequence, including the presence of a significant number of AGN-dominated systems prior to coalescence and starburst-dominated but fully merged systems, implies that black hole accretion, in addition to depending on the merger phase, also has a strong chaotic/random component, as in local AGNs.