2010ApJ...714..367Z

We present a sample of 68 low-z Mg II low-ionization broad absorption-line (loBAL) quasars. The sample is uniformly selected from the Sloan Digital Sky Survey Data Release 5 according to the following criteria: (1) redshift 0.4 < z ≤ 0.8, (2) median spectral S/N>7/pixel, and (3) Mg II absorption-line width Δv_c≥ 1600 km/s. The last criterion is a trade-off between the completeness and consistency with respect to the canonical definition of BAL quasars that have the "balnicity index" BI>0 in C IV BAL. We adopted such a criterion to ensure that ∼90% of our sample are classical BAL quasars and the completeness is ∼80%, based on extensive tests using high-z quasar samples with measurements of both C IV and Mg II BALs. We found (1) Mg II BAL is more frequently detected in quasars with narrower Hβ emission line, weaker [O III] emission line, stronger optical Fe II multiplets, and higher luminosity. In term of fundamental physical parameters of a black hole accretion system, loBAL fraction is significantly higher in quasars with a higher Eddington ratio than those with a lower Eddington ratio. The fraction is not dependent on the black hole mass in the range concerned. The overall fraction distribution is broad, suggesting a large range of covering factor of the absorption material. (2) [O III]-weak loBAL quasars averagely show undetected [Ne V] emission line and a very small line ratio of [Ne V] to [O III]. However, the line ratio in non-BAL quasars, which is much larger than that in [O III]-weak loBAL quasars, is independent of the strength of the [O III] line. (3) loBAL and non-loBAL quasars have similar colors in near-infrared to optical band but different colors in ultraviolet. (4) Quasars with Mg II absorption lines of intermediate width are indistinguishable from the non-loBAL quasars in optical emission line properties but their colors are similar to loBAL quasars, redder than non-BAL quasars. We also discuss the implication of these results.