1: %\documentclass{aastex}
2: %\usepackage{emulateapj5,times,mathptm}
3: \documentclass[12pt,preprint]{aastex}
4: %\usepackage{emulateapj5}
5:
6: \shorttitle{}
7: \shortauthors{}
8:
9: \newcommand{\msun}{$M_{\odot}$}
10: \newcommand{\mbh}{$M_{\rm BH}$}
11: \newcommand{\ergs}{\rm erg\,s^{-1}}
12: \newcommand{\kms}{\rm km\,s$^{-1}$}
13: \newcommand{\SFR}{{\rm SFR}}
14: \newcommand{\myear}{M_\odot\, {\rm yr^{-1}}}
15: \newcommand{\feii}{\ion{Fe}{2}}
16: \newcommand{\FeIIHb}{\ion{Fe}{2}\,$\lambda$4570/H$\beta$}
17: \newcommand{\OIIItwo}{[\ion{O}{3}]\,$\lambda\lambda$4959,5007}
18: \def\sun{\hbox{$\odot$}}
19:
20: \begin{document}
21:
22: \title{THE PHYSICAL CONNECTIONS AMONG IR QSOs, PG QSOs AND NARROW-LINE SEYFERT 1 GALAXIES}
23:
24: \author{
25: C.N. Hao,\altaffilmark{1}
26: X.Y. Xia,\altaffilmark{2,1}
27: Shude Mao,\altaffilmark{3,2}
28: Hong Wu,\altaffilmark{1}
29: and Z.G. Deng\altaffilmark{4,1}}
30:
31: %
32:
33: \altaffiltext{1}{National Astronomical Observatories,
34: Chinese Academy of Sciences, A20 Datun Road, 100012 Beijing,
35: China; Email: hcn@bao.ac.cn.}
36: \altaffiltext{2}{Dept. of Physics, Tianjin Normal University,
37: 300074 Tianjin, China.}
38: \altaffiltext{3}{Univ. of Manchester, Jodrell Bank Observatory,
39: Macclesfield, Cheshire SK11 9DL, UK.}
40: \altaffiltext{4}{College of Physical Science, Graduate School of
41: the Chinese Academy of Sciences, 100039 Beijing, China.}
42:
43: \received{\date}
44: \accepted{}
45:
46: \begin{abstract}
47:
48: We study the properties of infrared-selected QSOs (IR QSOs),
49: optically-selected QSOs (PG QSOs) and Narrow Line Seyfert 1 galaxies (NLS1s).
50: We compare their properties from the infrared to the optical and
51: examine various correlations among
52: the black hole mass, accretion rate, star formation rate and
53: optical and infrared luminosities. We find that
54: the infrared excess in IR QSOs is mostly in the far infrared, and their
55: infrared spectral indices suggest that the excess emission is from
56: low temperature dust heated by starbursts rather than AGNs. The
57: infrared excess is therefore a useful criterion to separate the
58: relative contributions of starbursts and AGNs.
59: We further find a tight correlation between the star formation rate
60: and the accretion rate of central AGNs for IR QSOs. The ratio
61: of the star formation rate and the accretion rate
62: is about several hundred for IR QSOs, but decreases with the central
63: black hole mass. This shows that the tight correlation between the stellar
64: mass and the central black hole mass is preserved in massive
65: starbursts during violent mergers. We suggest that
66: the higher Eddington ratios of NLS1s and IR QSOs
67: imply that they are in the early stage of evolution
68: toward classical Seyfert 1's and QSOs, respectively.
69: \end{abstract}
70:
71: \keywords{galaxies: active --- galaxies: evolution ---
72: galaxies: interactions --- galaxies: ISM --- quasars: general --- galaxies: starburst}
73:
74:
75: \section{INTRODUCTION}
76:
77: Although much effort has been made since the local ultraluminous
78: IRAS galaxies (ULIGs) were discovered, the dominant energy output
79: mechanism and the evolutionary connection between
80: circum-nuclear starbursts and active galactic nuclei (AGNs)
81: are still a matter of debate
82: (Kim, Veilleux, \& Sanders 1998; Veilleux, Kim, \& Sanders 1999a; Veilleux,
83: Sanders, \& Kim 1997, 1999b; Goldader et al. 1995; Murphy et al. 2001;
84: Genzel et al. 1998; Lutz et al. 1998; Rigopoulou et al. 1999; Tran et al. 2001).
85: From recent high resolution multi-wavelength observations,
86: it is now widely accepted that the vast majority of ULIGs
87: are strongly interacting or merging galaxies (e.g., Clements et al. 1996;
88: Murphy et al. 1996; Veilleux, Kim, \& Sanders 2002). The AGN
89: phenomenon appears at the final merging stage
90: and the fraction of objects with AGN spectral characteristics
91: is about 30\% while the fraction of type 1 AGNs is less than 10\%
92: (e.g., Clements et al. 1996; Kim, Veilleux, \& Sanders 1998; Wu et al. 1998;
93: Zheng et al. 1999; Canalizo \& Stockton 2001; Cui et al. 2001).
94: However, the percentage of AGNs increases with
95: infrared luminosity, reaching 30--50\% for $L_{\rm IR} > 10^{12.3}L_{\odot}$
96: (Veilleux, Kim, \& Sanders 1999a).
97:
98: Zheng et al. (2002) carefully investigated the optical
99: spectroscopic properties of infrared-selected type 1 AGNs consisting of 25
100: objects with infrared luminosities $L_{\rm IR} > 10^{12}L_{\odot}$.
101: (Following Zheng et al. 2002, we refer to them as IR QSOs.)
102: They found that the majority of IR QSOs have relatively narrow
103: permitted emission lines compared with optically-selected Palomar Green quasars
104: (PG QSOs), which are from the Palomar Bright Quasar Survey Catalogue (Schmidt \& Green 1983)
105: with redshift less than 0.5 (Boroson \& Green 1992, hereafter BG92).
106: Furthermore, more than 70\% of IR QSOs are moderately or extremely strong \feii\
107: emitters. Canalizo \& Stockton (2001) proposed that such IR QSOs are at
108: a transitional stage between ULIGs and optically-selected QSOs as their host galaxies
109: are undergoing interacting or major merging accompanied
110: by massive starbursts. In fact, all the optical
111: spectroscopic properties of IR QSOs show that they
112: are located at one extreme end of Eigenvector 1 (or the first Principal
113: Component) defined by BG92 in their principal
114: component analysis. As pointed out by Grupe (2004), Eigenvector 1
115: correlates well with the Eddington luminosity ratio $L$/${L_{\rm Edd}}$.
116: This ratio is thought to be indicative of the `age' of an AGN --
117: AGNs with a higher Eddington luminosity ratio are at the onset of an
118: AGN phase. Important clues can therefore be gathered by studying
119: young forming QSOs with massive starbursts in order to
120: understand the physics of merging galaxies
121: and the AGN phenomenon. Moreover, the co-existence of starbursts and AGNs
122: provides important information about
123: the buildup of the stellar populations in
124: galaxies and the growth of central black holes.
125:
126: As mentioned above, the fraction of objects with AGN spectral
127: characteristics is about 30\% among ULIGs.
128: High resolution X-ray observations by Chandra and XMM-Newton
129: confirmed the existence of central AGNs in some ULIGs through the detection
130: of ${\rm Fe\, K}\alpha$ lines, for example in NGC\,6240, Mrk\,273, Mrk\,231 and IRAS\,19254-7545
131: (Xia et al. 2002; Komossa et al. 2003; Franceschini et al. 2003).
132: However, even for AGNs among ULIGs, the dominant energy output
133: is probably massive starbursts, instead of a central AGN engine.
134: For example, using the CO kinematic data of IR QSO Mrk\,231,
135: Downes \& Solomon (1998) concluded that the central AGN provides only
136: one third of the total luminosity, with the rest contributed by starbursts.
137:
138: Massive starbursts may dominate the energy output not only
139: for some local IR QSOs but also for high redshift massive
140: starburst galaxies and some optically-selected QSOs.
141: A recent deep SCUBA survey uncovered a large population of ULIGs at $z>1$
142: (Wang et al. 2004 and references therein).
143: Based on ultra-deep X-ray observations and deep optical spectroscopic data,
144: Alexander et al. (2004) argued that about 40\% of bright SCUBA sources
145: host AGNs. However, only $\la 20\%$ of the bolometric luminosity
146: is contributed by AGNs. Recently, Carilli et al. (2004) reported that
147: 30\% of optically-selected QSOs at high redshift are hyper-luminous
148: far-infrared galaxies with
149: $L_{\rm IR} > 10^{13}L_{\odot}$ and with dust masses $>10^{8}M_{\odot}$.
150: These QSOs follow the radio to far-infrared correlation for star-forming
151: galaxies (Carilli et al. 2001). Therefore, the main energy source for
152: the SCUBA detected AGNs and some high redshift optical QSOs
153: may also be starbursts. As the comoving luminosity density of infrared light contributed
154: by luminous infrared galaxies at $z \sim 1$ is more than 40 times larger than that
155: in the local universe (Elbaz et al. 2002), it is important to
156: investigate the properties of these
157: objects in order to understand the star formation history of the universe
158: and the number counts of AGNs at high redshift (Alexander et al. 2004).
159: Lessons we learn on how to determine the
160: dominant energy output mechanism for local IR QSOs
161: will provide clues to understanding the nature of AGNs
162: at higher redshifts and the processes involved in
163: galaxy formation and evolution.
164:
165: In this paper, we perform statistical analyses for
166: IR QSOs and compare their properties with
167: those of PG QSOs and narrow-line Seyfert 1 galaxies (NLS1s).
168: The outline of the paper is as follows. In \S 2,
169: we describe how the IR QSO, PG QSO and NLS1 samples are
170: compiled. In \S 3, we discuss the data reduction and
171: how we estimate different physical parameters.
172: The statistical correlations are studied in \S 4. Finally,
173: in \S 5, we summarize and discuss our results. Throughout
174: this paper we adopt a cosmology with
175: a matter density parameter $\Omega_{\rm m}=0.3$, a cosmological constant
176: $\Omega_{\rm \Lambda}=0.7$ and
177: a Hubble constant of $H_{\rm 0}=70\,{\rm km \, s^{-1} Mpc^{-1}}$.
178:
179:
180: \section{SAMPLE SELECTION}
181:
182: One aim of our study is to understand the connection of star formation
183: and accretion process to the central black hole. For this
184: purpose, we use an infrared-selected type 1 AGN sample as
185: the star formation and AGN activity are coeval in these objects.
186: For comparison, we also compile an optically-selected QSO sample and a NLS1
187: sample, for which the infrared information is available.
188: The details of these three samples are given below:
189: \begin{enumerate}
190: \item[(1)] The infrared-selected type 1 AGN (IR QSO) sample is primarily from
191: Zheng et al. (2002). This sample was compiled from the ULIGs in the QDOT redshift
192: survey (Lawrence et al. 1999), the 1$\,$Jy ULIG survey (Kim \& Sanders 1998),
193: and an IR QSO sample obtained by a cross-correlation study of the IRAS Point-Source
194: Catalog with the ROSAT All-Sky Survey Catalog. All the IR QSOs selected
195: by Zheng et al. are ULIGs with mid-infrared
196: to far-infrared properties from IRAS observations.
197: These galaxies include most of the transition QSOs defined by
198: Canalizo \& Stockton (2001). Furthermore, they have also been carefully
199: investigated by Zheng et al. with optical spectra; they concluded
200: that these objects are in transition from ULIGs to classical QSOs or from
201: mergers to elliptical galaxies through a QSO phase.
202: We added three IR QSOs to the sample obtained from the cross-correlation
203: of the largest IRAS redshift survey (PSCz)
204: and the ROSAT archive by Xia et al. (2001). In total,
205: we have 28 objects, all of which are in the northern sky ($\delta > -30\degr$)
206: and they constitute about one third of all the IR QSOs identified using
207: PSCz, hence it should be a representative sample of IR QSOs.
208: %
209: %
210: \item[(2)] The optically-selected QSO sample comprises 57 PG QSOs from
211: the BG92 sample. For 51 of these,
212: their infrared information were taken from Haas et al. (2003).
213: We added six additional objects obtained by a cross-correlation of
214: 87 PG QSOs in BG92 and the IRAS Faint Source Catalog.
215: %
216: %
217: \item[(3)] A NLS1 sample was taken from Wang \& Lu (2001) with available IRAS
218: flux densities from the NED database\footnote{The NASA/IPAC Extragalactic Database (NED) is
219: operated by the Jet Propulsion Laboratory,
220: California Institute of Technology,
221: under contract with the National Aeronautics and Space Administration. }.
222: This sample is a heterogeneous sample compiled and observed
223: spectroscopically by Veron-Cetty et al. (2001). The sample consists of 39
224: objects, two of which overlap with
225: the IR QSO sample and eight overlap with the PG QSO sample. In
226: addition, the $M_{\rm BH}$ estimation is not available for one of them (MS\,15198$-$0633), so we
227: excluded this object from the NLS1 sample, leaving a total of 28 objects.
228: \end{enumerate}
229:
230: One complication we already alluded to
231: is that there are overlapping objects among the three samples.
232: For later statistical analyses, we re-group them as follows:
233: (1) The overlapping objects between the IR QSO and PG QSO samples
234: are classified in the IR QSO sample. Furthermore, three PG QSOs
235: (PG\,0050$+$124, PG\,1543$+$489 and PG\,1700$+$518) are re-classified
236: as IR QSOs following Canalizo \& Stockton (2001) who defined these as in
237: transition from ULIGs to classical QSOs\footnote{Canalizo \& Stockton
238: (2001) compiled a sample of transition objects
239: with nine QSOs and six other similar objects, which were included in the sample
240: of Zheng et al. (2002).}.
241: (2) The original PG QSO sample was selected optically, regardless of the FWHM
242: of H$\beta$ emission line. In fact, some of the PG QSOs have
243: FWHM of H$\beta$ less than 2000\kms, satisfying the criteria of
244: NLS1s\footnote{These PG QSOs were classified as NLS1s in the literature as well.}.
245: We will therefore put these objects into the NLS1 sample.
246: After this exercise, the numbers of objects in the IR QSO, PG
247: QSO and NLS1 samples are 31, 41 and 38, respectively.
248:
249: One criterion of our sample selection is that the objects must
250: have infrared flux or luminosity information. As the infrared
251: information is not as readily available as the optical information,
252: all the three samples used here are somewhat incomplete. However, they
253: are representative of IR QSOs, optically-selected classical QSOs and NLS1s
254: in the local universe. Furthermore, as discussed in \S 1,
255: there are optical spectroscopic similarities
256: between IR QSOs and NLS1s and possible evolutionary connections between IR QSOs
257: and optically-selected QSOs (Sanders et al. 1988a). Therefore,
258: these three samples, while incomplete, will allow us to explore the physical
259: relations among these three classes of objects.
260:
261: \section{DATA REDUCTION AND ESTIMATION OF PHYSICAL PARAMETERS}
262:
263: In this section, we briefly discuss the data reduction and describe how
264: we determine the physical parameters of AGNs,
265: including their black hole masses, infrared and optical luminosities,
266: H$\beta$ Luminosities,
267: star formation rates and accretion rates.
268: All the parameters are listed in Table 1.
269:
270: \subsection{Black Hole Masses \label{sec:bh}}
271:
272: The method used to estimate a black hole mass is based on the assumption that
273: the motion of the gas moving around the black hole is dominated by the
274: gravitational force and
275: the broad emission line region (BLR) gas is virialized (see Peterson \& Wandel
276: 1999, 2000 for evidence). Hence the central black hole mass can be estimated
277: using the BLR radius and velocity of the BLR gas, i.e.,
278: \begin{equation}\label{eq:Mbh}
279: M_{\rm BH} = {R_{\rm BLR} V^2 \over G},
280: \end{equation}
281: where ${G}$ is the
282: gravitational constant. The size of the BLR ($R_{\rm BLR}$) can be estimated from the empirical
283: relationship between the size and the monochromatic continuum
284: luminosity at 5100\AA.
285: This relation was first found by Kaspi et al. (2000) for a sample of 17 Seyfert
286: 1 galaxies and 17 PG QSOs in a cosmology
287: with $H_{\rm 0}=75 {\rm km s^{-1} Mpc^{-1}}$, $\Omega_{\rm m}=1$, and
288: $\Omega_\Lambda=0$.
289: The relation was refitted in our adopted cosmology by
290: McLure \& Jarvis (2002)
291: %
292: \begin{equation} \label{eq:rBLR}
293: R_{\rm BLR} = (26.4\pm4.4)\left[\frac{\lambda L_{\lambda}(5100{\mbox{\AA}})}{10^{44}\,{\rm erg\,s}^{-1}}\right]^{(0.61\pm0.10)} \mbox{lt-days}.
294: \end{equation}
295: %
296: The velocity $V$ can be estimated from the Full Width at Half Maximum (FWHM) of the
297: H$\beta$ broad emission line
298: $V = \sqrt{3}/2V_{\rm FWHM}$, by assuming that the BLR gas is in
299: isotropic motions. Therefore, two measurements are needed to
300: determine the black hole
301: mass: the luminosity at 5100\AA\ and the FWHM of H$\beta$.
302:
303: We obtained the FWHM of H$\beta$ from Zheng et al. (2002) for 25 out of
304: the 28 IR QSOs. For the remaining three IR QSOs (IRAS\,F01348$+$3254,
305: IRAS\,03335$+$4729 and IRAS\,F04505$-$2958),
306: we observed them and reduced the
307: spectra in the same manner as described by Zheng et al. (2002).
308: Notice that the FWHM of H$\beta$ was estimated in the same way as in BG92.
309: The continuum flux densities at 5100\AA\ were measured directly from our
310: spectra. The uncertainties of the black hole mass were estimated by
311: error propagation using the uncertainties of the flux density and the FWHM of H$\beta$
312: measurements given by Zheng et al. (2002). The mean error of the black hole mass
313: is 0.13dex. This is a lower limit as the uncertainties
314: of FWHM of H$\beta$ given by Zheng et al. are probably under-estimated (cf. Shemmer et al. 2004)
315: and there exists other sources of systematic errors (see Wang \& Lu 2001 for more detailed
316: discussions). Generally, the black hole mass derived in this way is
317: accurate within a factor of 2$-$3 (e.g., Wang \& Lu 2001; Marziani et al. 2003, and references
318: therein; Shemmer et al. 2004).
319:
320: For the 57 PG QSOs in our sample, the FWHM of H$\beta$ measurements were
321: from BG92.
322: The continuum flux densities at 5100\AA\ were taken from the spectrophotometry by
323: Neugebauer et al. (1987). Specifically, the flux
324: densities at 5100\AA\ were estimated by a linear interpolation over the
325: neighboring frequency range.
326:
327: For the 28 NLS1s, we used the BLR sizes listed by Wang \& Lu
328: (2001) to calculate the monochromatic luminosity at 5100\AA\
329: using
330: the $R_{\rm BLR}-\lambda L_{\rm 5100}$ relation given by Kaspi et al. (2000),
331: then used eq. (\ref{eq:rBLR}) and
332: eq. (\ref{eq:Mbh}) to
333: derive the black hole masses from the H$\beta$ FWHM and monochromatic luminosity at 5100\AA.
334: In this process, we carefully accounted for the difference in the adopted cosmology
335: (Wang \& Lu 2001 used a cosmology with $H_0=75{\rm km\,s^{-1}}\,{\rm Mpc}^{-1}$,
336: $\Omega_m=1$ and $\Omega_\Lambda=0$).
337:
338:
339: \subsection{Infrared Luminosities}
340:
341: For all the sample objects except the 51 PG QSOs, we calculated their infrared
342: luminosities following Sanders \& Mirabel (1996) based on the flux
343: densities from the IRAS Faint Source Catalog:
344: \begin{equation}
345: L(8-1000\mu {\rm m})=4\pi D_L^2 f_{\rm IR},
346: \end{equation}
347: where $D_L$ is the luminosity distance,
348: and $f_{\rm IR}$ is defined as
349: \begin{equation}
350: f_{\rm
351: IR}=1.8\times10^{-14}\{13.48f_{12}+5.16f_{25}+2.58f_{60}+f_{100}\} {\rm W\,m^{-2}}
352: \end{equation}
353: with $f_{12}$, $f_{25}$, $f_{60}$ and $f_{100}$ being the IRAS flux densities at
354: 12, 25, 60 and 100$\mu{\rm m}$ in units of Jy.
355:
356: Notice that three IR QSOs (IRAS\,06269$-$0543, IRAS\,11598$-$0112 and IRAS\,03335$+$4729)
357: were not in the IRAS Faint Source Catalog, so we obtained their flux densities from the
358: IRAS Point Source Catalog. The typical uncertainty of infrared luminosities
359: is about 0.06 dex. For all the PG QSOs except the six
360: objects (i.e., PG\,0923$+$129, PG\,0923$+$201, PG\,1119$+$120, PG\,1351$+$236, PG\,1534$+$580 and PG\,1612$+$261)
361: obtained by cross-correlating 87 PG QSOs and the IRAS Faint Source Catalog,
362: their infrared luminosities were calculated by summing over the
363: $L_{\rm NIR}(3-10\mu {\rm m})$,
364: $L_{\rm MIR}(10-40\mu {\rm m})$, $L_{\rm FIR}(40-150\mu {\rm m})$ (from ISO observations) and
365: $L_{\rm sub-mm}(150-1000\mu {\rm m})$ (from the MAMBO and SCUBA
366: (sub-)millimeter data) in Table 2 of Haas et al. (2003).
367: There are 16 common objects in the IRAS Faint Source Catalog and Haas et al. (2003).
368: The average difference of the infrared luminosities measured from the IRAS Faint Source Catalog
369: and Table 2 of Haas et al. (2003) for these galaxies
370: is 0.014 dex. Hence the infrared luminosities derived in these two different ways
371: agree well and will not lead to large systematic errors.
372:
373: Monochromatic luminosities ($\lambda L_\lambda=\nu L_\nu$) at $12\mu{\rm m}$, $25\mu{\rm m}$, $60\mu{\rm m}$ and
374: $100\mu{\rm m}$ were also calculated.
375: The flux densities in the four bands of the 51 PG QSOs were
376: derived from Table 1 of Haas et al. (2000, 2003). Notice that some PG
377: QSOs do not have information in all
378: four bands\footnote{PG\,1352+183 has no $12\mu{\rm m}$ observation;
379: PG\,0007$+$106, PG\,1259$+$593, PG\,1302$-$102, PG\,1307$+$085, PG\,1352$+$183,
380: PG\,1411$+$442, PG\,1425$+$267 and PG\,2112$+$059 have no $25\mu{\rm m}$ observations;
381: PG\,1425+267 has no $60\mu{\rm m}$ observation; PG\,1425$+$267 and PG\,1545$+$210 have no
382: $100\mu{\rm m}$ observations.}. For most objects observed by both IRAS and ISO,
383: the difference of the flux densities at $12\mu{\rm m}$, $25\mu{\rm m}$ and $60\mu{\rm m}$ is within
384: 30\%. However, the difference of flux densities at $100\mu{\rm m}$ is larger
385: than 30\% for half of the objects.
386: This could be due to the large IRAS beam of about
387: 3$^{'}$ which may enclose cirrus contaminations in
388: the flux determinations
389: (Haas et al. 2003). The difference of flux densities at
390: $12\mu{\rm m}$, $25\mu{\rm m}$, $60\mu{\rm m}$ and $100\mu{\rm m}$ are 0.072dex, 0.021dex, 0.002dex
391: and 0.225dex, respectively. However, the larger
392: errors at $100\mu{\rm m}$ should not statistically
393: affect our results that are mainly based on the
394: flux densities at $60\mu{\rm m}$. Note that no k-correction\footnote{
395: The k-correction here is defined as the ratio of the intrinsic luminosity
396: and the observed luminosity.} is applied
397: to any of the monochromatic or integrated infrared luminosities.
398: We estimated the k-correction by assuming our objects follow the
399: spectral energy distribution of either luminous infrared AGNs
400: or infrared starburst galaxies from Xu et al. (2001). We calculated the k-correction
401: from $z=0.$ to $z=0.5$ and
402: found that the largest k-correction occurs for a starburst galaxy
403: at $z=0.5$. Even in this case, the k-correction is
404: smaller than 0.25dex for the monochromatic luminosity at $25\mu{\rm m}$.
405: As all our sample objects have redshifts smaller than 0.5 and many of them
406: have spectral energy distributions similar to AGNs, the k-correction
407: is much smaller than 0.25dex. Hence our results
408: are not affected significantly by k-corrections.
409:
410: \subsection{Optical and H$\beta$ Luminosities}
411:
412: The optical luminosities of central AGNs of IR QSOs and PG QSOs were estimated by
413: the monochromatic continuum luminosity at 5100\AA,
414: $L_{\rm opt}=\lambda L_{\lambda}(5100{\mbox{\AA}})$. For PG QSOs, this
415: approach appears reasonable, as $L_{\rm opt}$'s determined in this way are
416: consistent with the $V$-band luminosities obtained through photometric observations.
417: We will discuss the origin of the optical continuum and the
418: applicability of this formula to IR QSOs in more detail in \S 4.1.
419: For IR QSOs, the typical uncertainty of optical luminosities derived in
420: this way is about 10\%--20\%.
421:
422: For NLS1s, we used the BLR sizes listed by Wang \& Lu (2001) and
423: the $R_{\rm BLR}-\lambda L_{\rm 5100}$ relation given by
424: Kaspi et al. (2000, cf. eq. \ref{eq:rBLR}) to derive the optical luminosities
425: of central AGNs (see the end of \S3.1).
426:
427: H$\beta$ luminosities were estimated by the H$\beta$ equivalent widths
428: and the
429: monochromatic continuum luminosities at 5100\AA.
430: The equivalent widths of the
431: broad H$\beta$ component for NLS1s were obtained from Veron-Cetty et al. (2001).
432: For IR QSOs and PG QSOs, the equivalent widths of H$\beta$ were from
433: Zheng et al. (2002) and BG92, respectively. Notice that
434: for IR QSOs and PG QSOs, the
435: equivalent width of H$\beta$ includes a contribution from a narrow
436: H$\beta$ component, but as BG92 and Zheng et al. (2002) discussed, this
437: contribution is less than 3\% of the total H$\beta$ flux,
438: so the narrow component will not affect the results significantly.
439:
440: \subsection{Bolometric Luminosities and Eddington Ratios}
441:
442: For IR QSOs, PG QSOs and NLS1s, we estimated the bolometric luminosities
443: using (Kaspi et al. 2000),
444: \begin{equation} \label{eq:Lbol}
445: L_{\rm bol}\approx9\lambda L_{\lambda}(5100{\rm \mbox{\AA}}).
446: \end{equation}
447: Notice that the bolometric luminosity
448: only refers to the total luminosity associated with the central AGN.
449:
450: The Eddington ratio $L_{\rm bol}/L_{\rm Edd}$ was then
451: derived, where the Eddington
452: luminosity was calculated using the black hole masses
453: (e.g., Peterson 1997) already determined in \S\ref{sec:bh}.
454:
455: \subsection{Accretion Rates and Star Formation Rates}
456:
457: The accretion rate onto the black hole was calculated by
458: \begin{equation}
459: L_{\rm bol}=\eta\dot{M} c^2,
460: \end{equation}
461: where $\eta$ is the accretion efficiency and
462: $\dot{M}$ is mass accretion rate. We adopt
463: $\eta=0.1$ throughout this paper (as in many other papers), and we have
464: \begin{equation}
465: \dot{M}=6.74\myear {L_{\rm bol} \over 10^{13}L_\odot}.
466: \label{mdoteq}
467: \end{equation}
468: The star formation rate was calculated by the monochromatic
469: luminosity at $60\mu {\rm m}$ due to starbursts (see \S 4.3)
470: through the following intermediate steps. Notice
471: that the $60\mu{\rm m}$ flux densities were used for this purpose because
472: they have the fewest upper limits.
473:
474: Lawrence et al. (1989) and Cardiel et al. (2003) gave, respectively,
475: \begin{equation}
476: L(40-120\mu{\rm m})\approx 2L_{60\mu{\rm m}}, ~~~
477: L(8-1000\mu{\rm m}) \approx (1.89\pm0.26)L(40-120\mu{\rm m}).
478: \end{equation}
479: So we have
480: \begin{equation}
481: L(8-1000\mu{\rm m})\approx 3.78L_{60\mu{\rm m}}.
482: \label{l60eq}
483: \end{equation}
484: Kennicutt (1998) gave an empirical calibration between the star
485: formation rate and $L(8-1000\mu{\rm m})$:
486: \begin{equation}
487: \SFR \approx 4.5 \,\myear {{L}(8-1000\mu{\rm m}) \over 10^{44} {\rm erg\,s^{-1}}}.
488: \label{sfrcon}
489: \end{equation}
490: Combining Eqs. (\ref{l60eq}) and (\ref{sfrcon}), we have
491: \begin{equation}
492: \SFR \approx 6.52\, \myear {{L}_{60\mu{\rm m}}\over 10^{10}L_\odot}.
493: \label{eq:sfreq}
494: \end{equation}
495:
496:
497: \section{RESULTS}
498:
499: Our three samples allow us to investigate the differences of physical properties
500: for IR QSOs, optically-selected bright PG QSOs and NLS1s and probe
501: possible evolutionary connections among these
502: objects, and, equally importantly, the interplay between starbursts and AGNs.
503:
504: We study the correlations between different
505: quantities for IR QSOs, PG QSOs and NLS1s.
506: For this purpose, we performed
507: Spearman Rank-order (S-R) correlation analyses. For some objects,
508: IRAS/ISO observations have only provided upper limits
509: of flux densities in one or two bands. When such data are present
510: we performed survival analysis\footnote{ASURV, Isobe, Feigelson \& Nelson (1986).}.
511: The correlation results are listed in Table 2. We discuss
512: these correlations in more detail below.
513:
514: \subsection{Correlations\label{sec:correlations}}
515:
516: Fig.~\ref{lhblirlop.eps}a shows that the featureless optical
517: continuum luminosity at 5100\AA\ correlates tightly with the broad H$\beta$ luminosity for
518: all three samples as a whole. This well-established relation
519: (Yee 1980; Shuder 1981, see also Osterbrock 1989)
520: is often used to argue that the predominant mechanism of BLR gas
521: excitation in AGNs is photo-ionization by the nuclear continuum (e.g., Veilleux,
522: Kim, \& Sanders, 1999a). Therefore, Fig.~\ref{lhblirlop.eps}a
523: suggests that central AGNs
524: power the optical emission for IR QSOs, PG QSOs and NLS1s.
525: This result is consistent with Barthel (2001) who concluded that for QSOs,
526: the $B$-band magnitude measures the AGN strength. Kauffmann \& Heckman (2004)
527: also pointed out that the optical continuum of type 1 AGN is dominated by
528: non-thermal emission. Therefore, it appears reasonable
529: to adopt a common relation (eq. \ref{eq:Lbol})
530: to estimate the bolometric luminosities from the central AGN for all the
531: objects.
532:
533: Fig.~\ref{lhblirlop.eps}b
534: shows the infrared luminosity versus the optical luminosity
535: for PG QSOs, NLS1s and IR QSOs. The solid line shows
536: the regression line between these two quantities
537: for the PG QSOs and NLS1s. One sees that a considerable fraction of IR QSOs
538: are above the line. If the tight
539: correlation between the infrared and the optical luminosity for PG QSOs and
540: NLS1s is because they are both associated with
541: central AGNs, then a reasonable
542: extrapolation for the infrared excess of IR QSOs is that there is
543: another energy source in addition to the AGN that heats the dust.
544:
545: To further clarify the mechanism that leads to the infrared excess for IR QSOs,
546: Fig.~\ref{l4band.lop.eps} shows the infrared emission in
547: four bands (12$\,\mu{\rm m}$, 25$\,\mu{\rm m}$, 60$\,\mu{\rm m}$ and 100$\,\mu{\rm m}$)
548: with the optical luminosity. Comparing the four panels,
549: one can see that the mid-infrared luminosity (at 12$\,\mu{\rm m}$, 25$\,\mu{\rm m}$)
550: correlates tightly with the optical luminosity for PG QSOs and
551: NLS1s, implying that the mid-infrared emission for optically-selected QSOs
552: and NLS1s is associated with a central AGN.
553: The correlations for the far-infrared luminosity (at 60$\,\mu{\rm m}$ and 100$\,\mu{\rm m}$) and optical luminosity
554: for PG QSOs and NLS1s are also tight, but
555: some PG QSOs show deviations (see
556: Fig.~\ref{l4band.lop.eps}d). Fig.~\ref{l4band.lop.eps} clearly shows
557: that IR QSOs deviate from the regression line for PG QSOs and NLS1s
558: in all four infrared bands; the deviations
559: are more dramatic in the far-infrared (panels c and d) than
560: those in the mid-infrared bands (panels a and b).
561: Therefore, the infrared excess of IR QSOs occurs
562: mainly in the far-infrared. This is
563: further supported by ISO observations which showed
564: that both PG QSOs and IR QSOs follow similar power-law spectral energy
565: distributions from the near-infrared
566: to mid-infrared band (Haas et al. 2003; Peeters et al. 2004).
567: The spectral similarities in the
568: mid-infrared suggest that the most significant difference
569: between PG QSOs and IR QSOs must occur in the far-infrared.
570:
571: The left panels of Fig.~\ref{his2panel1.eps} show the histogram of $L_{\rm IR}/L_{\rm bol}$
572: for different samples. As can be seen,
573: the PG QSOs and NLS1s have similar distributions;
574: their median values of $L_{\rm IR}/L_{\rm bol}$ are almost the same
575: ($\approx 0.33$), if we assume all the upper limits are real detections.
576: It implies that roughly one third of the bolometric
577: luminosity of the optically-selected type 1 AGNs is emitted in the infrared from their dust tori.
578: This is consistent with Sanders et al. (1989) who found that all PG QSOs
579: emit a significant fraction, 10\%--50\% with a typical value of 30\%, of their
580: bolometric luminosity in the infrared. In sharp contrast,
581: more than two thirds of IR QSOs have $L_{\rm IR}/L_{\rm bol}$ ratio larger
582: than one with a median value of 1.4, highlighting again
583: the significant infrared excess of the IR QSOs
584: compared with PG QSOs and NLS1s.
585:
586: Given that the infrared excess is mainly in the far-infrared,
587: we further examine the ratio of the monochromatic luminosity at $60\mu{\rm m}$ to the
588: bolometric luminosity.
589: The right panels of Fig.~\ref{his2panel1.eps} show the histograms of this ratio for
590: IR QSOs, PG QSOs and NLS1s. As can be seen, the median
591: value (0.50) of $L_{\rm 60 \mu{\rm m}}$/$L_{\rm bol}$ for
592: IR QSOs is significantly larger than those (0.09, 0.08) for PG QSOs and
593: NLS1s, confirming that IR QSOs have high far-infrared excesses compared
594: with optical QSOs and NLS1s.
595:
596: \subsection{Eddington Ratios and Spectral Indices}
597:
598: To investigate the physical connections of IR QSOs,
599: optically-selected QSOs and NLS1s, in the following we study
600: the Eddington accretion ratio, black hole mass and the relation
601: of infrared color with infrared excess ($L_{\rm IR} / L_{\rm bol}$).
602:
603: As is well known, the central engine of an AGN is powered
604: by matter accretion onto the black hole. The AGN luminosity is proportional
605: to both the mass accretion rate and the accretion efficiency, which is
606: determined by complex accretion physics. Assuming a fixed
607: accretion efficiency, the Eddington ratio, $L_{\rm bol} /L_{\rm Edd}$,
608: measures the accretion rate in units of the critical Eddington value.
609:
610: The left panels of Fig.~\ref{his2panel2.eps} show the histograms of $L _{\rm bol}/L_{\rm Edd}$
611: for three samples. Clearly the IR QSOs have a similar distribution of $L _{\rm bol}/L_{\rm Edd}$ as NLS1s.
612: The median values of the Eddington ratio are about 1.7 for IR QSOs and 1.3
613: for NLS1s, respectively. More than half of the
614: IR QSOs and NLS1s have $L_{\rm bol}/L_{\rm Edd}>1$,
615: implying that the accretion in these systems
616: may not be spherically symmetric (Collin et al. 2002; Wang 2003).
617: On the other hand, most PG QSOs have Eddington ratios smaller than one with
618: a median value of about 0.2, much smaller than those of IR QSOs and NLS1s.
619: As mentioned earlier, IR QSOs and NLS1s have similar optical spectroscopic
620: properties, and they lie
621: at one extreme end of the first Principal Component defined by BG92 (Zheng et al. 2002).
622: This component has been suggested to correlate with $L _{\rm bol}/L_{\rm Edd}$
623: (e.g., BG92; Shemmer \& Netzer 2002).
624: Therefore the spectroscopic similarities between IR QSOs and NLS1s may
625: be due to the fact
626: that both classes of objects have a high Eddington ratio.
627:
628: Next we investigate the black hole mass distribution
629: (right panels of Fig.~\ref{his2panel2.eps}). It is clear that
630: NLS1s have much smaller black hole masses compared with IR QSOs and PG QSOs.
631: The median black hole masses are $5 \times 10^{7}M_\odot$,
632: $2 \times 10^{8}M_\odot$ and $6.5 \times 10^{6}M_\odot$
633: for IR QSOs, PG QSOs and NLS1s, respectively. In addition, the distribution
634: of black hole mass of IR QSOs is broader than those of PG QSOs and NLS1s,
635: which can be explained if
636: the black hole masses of IR QSOs are still increasing and have
637: not yet reached a stable value.
638:
639: Fig.~\ref{alpha.lirlbol.eps} shows the infrared spectral index of $\alpha (60,25)$
640: versus the infrared excess of $L_{\rm IR} / L_{\rm bol}$,
641: where the spectral index is defined as
642: \begin{equation}
643: \alpha(\lambda_1, \lambda_2) =
644: -{\log(F(\lambda_2)/F(\lambda_1)) \over \log(\lambda_2/\lambda_1)}
645: \end{equation}
646: and the wavelengths are in units of $\mu{\rm m}$.
647: $\alpha(60,25)$ is a measure of the dust temperature (e.g., Sekiguchi 1987). The larger the
648: value, the higher the dust temperature.
649: There is a trend from Fig.~\ref{alpha.lirlbol.eps} that as $L_{\rm IR} / L_{\rm bol}$
650: decreases, $\alpha (60,25)$ increases, implying that the dust temperature increases
651: as the infrared excess decreases. Statistically, IR QSOs have lower $\alpha
652: (60,25)$ values and hence
653: lower dust temperature compared with optically-selected QSOs and
654: NLS1s. As the dust heated by AGNs tends to have higher temperatures,
655: this suggests that starbursts are
656: important for heating the dust in IR QSOs.
657: %
658: Our conclusions are supported by preliminary Spitzer
659: observations for IR QSO Mrk\,1014 (IRAS\,01572$+$0009, Armus et
660: al. 2004). These observations
661: clearly detected the 6.2, 7.7 and 11.3 $\mu{\rm m}$ PAH emission features,
662: demonstrating convincingly the existence of massive starbursts in IR QSOs.
663:
664:
665: \subsection{Star Formation Rates and Accretion Rates}
666:
667: From our discussions above, IR QSOs are accreting and forming stars at the
668: same time, below we investigate how these two processes are related to
669: each other.
670: %
671: Fig.~\ref{l4band.lop.eps}c is a plot of $L_{\rm 60 \mu{\rm m}}$ versus
672: the accretion rate
673: $\dot{M}$ for IR QSOs, PG QSOs and NLS1s.
674: From Fig.~\ref{l4band.lop.eps}c, it appears
675: that $L_{\rm 60 \mu{\rm m}}$ and $\dot{M}$ are correlated for
676: the IR QSO sample and the combined sample of
677: PG QSOs and NLS1s. However, these two sub-samples follow
678: different regression lines with quite different intercepts
679: (compare the dashed and solid lines in
680: Fig.~\ref{l4band.lop.eps}c).
681: As we discussed above, the far-infrared excess of IR QSOs are due to
682: the additional contribution of starbursts to the far-infrared luminosity
683: compared with PG QSOs and NLS1s.
684:
685: We can use Fig.~\ref{l4band.lop.eps}c to
686: estimate the star formation rate of IR QSOs.
687: As central AGNs also provide contribution to the
688: far-infrared emissions for IR QSOs, we first subtract this contribution by
689: assuming it follows the same
690: regression relation as PG QSOs and NLS1s (the solid curve in Fig.~\ref{l4band.lop.eps}c).
691: We then use the excess infrared luminosity at $60\mu{\rm m}$
692: to calculate the star formation rate using eq. (\ref{eq:sfreq}). We find that
693: the star formation rate and accretion rate are related to each other by
694: \begin{equation} \label{eq:sfr}
695: \log\, \SFR=(0.29\pm0.10)\log\dot{M} +(2.77\pm0.06),
696: \end{equation}
697: where $\SFR$ and $\dot{M}$ are both in units of $\myear$. The above
698: relation can be rewritten as
699: \begin{equation} \label{eq:sfrmdot}
700: {\SFR \over \myear}=588.8 \left({\dot{M} \over \myear}\right)^{0.29}.
701: \end{equation}
702: %
703: In order to examine whether a systematic trend exists between the star
704: formation rate, accretion rate and black hole mass, in
705: Fig.~\ref{ir.sbsfrmdot.mbh.eps} we plot
706: the ratio of the SFR to $\dot{M}$ versus the black hole mass for IR QSOs.
707: The regression line is given by
708: \begin{equation} \label{eq:sfrm}
709: \log {\SFR \over \dot{M}}=(-0.52\pm0.09)\log \left(M_{\rm BH}\over M_{\odot} \right)+(6.62\pm0.71).
710: \end{equation}
711: Eqs. (\ref{eq:sfrmdot}) and (\ref{eq:sfrm}) provide important clues about
712: how the star formation rate and the growth of black holes are connected in
713: violent merging galaxies. We return to this important
714: point in the discussion (\S 5.1).
715:
716:
717: \section{SUMMARY AND DISCUSSION}
718:
719: In this paper we have analyzed the statistical properties of IR QSOs, PG QSOs and NLS1s
720: from the optical to the infrared. Our results
721: reveal that these three classes of objects have distinct properties in the infrared.
722: Starbursts play a main role in the infrared
723: excess, especially the far-infrared excess, in IR QSOs.
724: Our study also reveals a correlation between the star formation rate
725: and the accretion rate onto central black holes during galaxy merging and
726: massive starbursts. This implies that the accretion-driven growth
727: of central black holes is correlated with the formation of young
728: stellar population. This has important
729: implications for the origin of the tight correlation between the
730: stellar
731: mass of the hot component of galaxies with the central black hole masses.
732: In the following we discuss these issues in more detail.
733:
734: \subsection{Correlation Between Star Formation Rates and Accretion Rates}
735:
736: In the last few years, it has become increasingly clear that the star formation
737: and AGN activity must be correlated as there are tight
738: correlations between the black hole mass,
739: galactic velocity dispersion (e.g. Ferrarese \& Merritt 2000) and
740: the mass or luminosity of the hot stellar component of the
741: host galaxy (e.g. Kormendy \& Gebhardt 2001; Magorrian et al. 1998; Laor
742: 1998). It is unclear how the correlations arise.
743:
744: By studying a large sample of narrow emission line galaxies
745: from the Sloan Digital Sky Survey (SDSS), Heckman et al. (2004; see also
746: Kauffmann \& Heckman 2004) found that the host galaxies of
747: bright AGNs have a much younger mean stellar age and quite a large fraction
748: of these host galaxies have experienced recent starbursts. More importantly,
749: they found that the volume averaged ratio of the star formation rate to
750: the black hole
751: accretion rate is about 1000 for bulge-dominated galaxies. This value
752: is in agreement with the ratio of the bulge mass to the
753: black hole mass empirically
754: derived (Marconi \& Hunt 2003). Notice that their results are
755: based on 23000 narrow emission-line galaxies which excludes type 1
756: AGNs. In their study, the black hole mass covers more than two orders of magnitude.
757: %
758: Our study, on the other hand, is based on only a few tens of
759: infrared-selected type 1 AGNs. But these objects are
760: experiencing both massive starbursts and rapid black hole growth
761: due to accretions at the same time, and so we probe
762: the same correlation but in more extreme environments.
763:
764: At this transitional stage from massive starburst to classical QSO,
765: the average ratio of star formation rate to black hole accretion rate
766: $\SFR / \dot{M}$ is also a few hundred as shown in Eq.~(\ref{eq:sfrmdot}).
767: Comparing Fig.~\ref{ir.sbsfrmdot.mbh.eps} with Figure 11 in Kauffmann \&
768: Heckman (2004), both the slope and zero-points of
769: the fitted lines are similar.
770: It is intriguing that $\SFR/\dot{M}$ is
771: not constant, but declines with the black hole mass; the same trend
772: was found in Kauffmann \& Heckman (2004).
773:
774: On the other hand, as we emphasized in \S4.2, the infrared emission, even
775: the far-infrared emission, for optically-selected QSOs and NLS1s are not
776: from the star formation, but mainly from dust heated by central
777: AGNs. For these objects, the accretion process is still powering
778: the AGN's emission, but there is no longer substantial
779: star formation. The picture we obtained here
780: is consistent with the
781: recent simulation result by Springel et al. (2004) that the starburst and AGN activity are coeval, but
782: the time durations are different as a result of
783: the detailed form of the response of the gas to the feedback processes.
784:
785: The number density of ULIGs and classical QSOs are comparable
786: in the local universe (Sanders et al. 1988a, 1988b; Canalizo \& Stockton 2001).
787: The fraction of IR QSOs is less than 10\% of ULIGs, hence
788: the number density of IR QSOs in the local universe is at most 10\%
789: of classical QSOs. If the number density of objects is simply related
790: to the time scale of different phases, then the time scale for IR QSOs
791: will be roughly 10\% of that of classical QSOs (about a few times
792: $10^{8}$ years, e.g., Marconi et
793: al. 2004). The IR QSO phase may therefore last only a few times $10^{7}$ years.
794: The co-moving number density of ULIGs is likely much higher at higher
795: redshift. For example,
796: Elbaz et al. (2002) found that the comoving number density of ULIGs is
797: several tens times higher at $z \sim 1$ than that in the local universe.
798: Correspondingly, the co-moving number density of IR QSOs can be higher by the same factor, i.e.,
799: the co-existing massive starbursts and rapid accretions onto
800: black holes may be much more common at higher redshift.
801: An investigation into the evolution of the co-moving number density of
802: IR QSOs with redshift will provide valuable information for
803: the formation of spheroidals and AGNs.
804:
805: \subsection{Infrared-Excess as Criterion of Starbursts}
806:
807: Based on an analysis for 64 PG QSOs from the infrared to the X-ray,
808: Haas et al. (2003) concluded that the central AGN is the dominant energy
809: source for all emissions of PG QSOs. Even for the far-infrared emission,
810: the central AGN is still the main source of heating on
811: the dust tori. If the dust
812: torus is clumpy, then the central AGN emission
813: can travel through the gaps farther out and provide the
814: observed far-infrared emission reradiated
815: from cooler outer regions. Therefore for PG QSOs, starbursts
816: play a minor role. In this paper, we showed that NLS1s
817: and PG QSOs also satisfy the same correlations and central AGNs
818: may be the dominant sources from the optical to the far-infrared,
819: even for some PG QSOs with $L_{\rm IR} > 10^{12} L_{\odot}$.
820:
821: The IR QSOs, PG QSOs and NLS1s
822: follow the same correlation between
823: the H$\beta$ luminosity and the optical continuum luminosity
824: (see Fig.~\ref{lhblirlop.eps}a and Table 2). This implies that for
825: all sample objects, regardless of whether they are optically-selected or
826: infrared-selected, their optical luminosity measures the
827: central AGN's power. In all the other correlations,
828: the IR QSOs are significantly different from the PG QSOs and NLS1s,
829: which is a direct result of the far-infrared excess in IR QSOs. We
830: showed that this infrared excess is from starbursts, as can be
831: most clearly seen from the larger values of
832: $L_{\rm IR}/L_{\rm bol}$ and smaller values of $\alpha (60,25)$ (see Fig.~\ref{alpha.lirlbol.eps}).
833: For most IR QSOs starbursts play a significant,
834: even dominating, role in their energy output. We conclude that the infrared excess
835: can serve as an efficient criterion to disentangle the relative
836: energy contributions of AGNs and starbursts.
837:
838: \subsection{Evolutionary Connections Between NLS1s, IR QSOs and PG QSOs}
839:
840: The optical spectra of IR QSOs and NLS1s are quite similar, both of which have strong or
841: extremely strong optical \feii\ emission and weak \OIIItwo\ emission.
842: They are located at one extreme end of the Eigenvector 1 of BG92.
843: It is widely accepted that the Eigenvector 1 is closely correlated with the
844: Eddington ratio ($L_{\rm bol} /L_{\rm Edd}$); a large Eddington ratio %
845: is interpreted as the AGN being at the early stage of an AGN phase,
846: i.e., having a young `age' (Grupe 2004). As both IR QSOs and NLS1s have high
847: Eddington ratios (see left panels of Fig.~\ref{his2panel2.eps}), it follows that both
848: classes of objects are young AGNs in their early stage of evolution.
849:
850: However, in all the analyses performed in section 4, NLS1s and PG QSOs
851: have similar
852: correlations. The differences between them are their
853: black hole masses and the Eddington ratios. On the other hand,
854: IR QSOs and NLS1s are different except that
855: both have high Eddington ratios. The difference between IR QSOs and
856: NLS1s is also
857: underlined by their host galaxies. It appears that the host galaxies of
858: IR QSOs are merging galaxies while the host galaxies of NLS1s are barred
859: spirals (Crenshaw et al. 2003). More importantly, IR QSOs are undergoing
860: massive starbursts which produce the infrared excess in these objects.
861:
862: The co-existing massive starbursts and high black hole accretion rate
863: in IR QSOs will lead to the rapid growth of
864: black holes and IR QSOs will rapidly evolve to classical QSOs hosted by elliptical galaxies.
865: While NLS1s have high Eddington accretion ratios,
866: the accretion rates $\dot{M}$ are in fact much smaller than those of IR QSOs
867: (see Fig.~\ref{l4band.lop.eps}c), therefore, the final black hole
868: for these objects will be smaller, and their
869: evolution destination may be Seyfert 1s, rather than QSOs.
870: This is also consistent
871: with the bulge and black hole mass relation --
872: the bulge mass of spiral galaxies is smaller than those in elliptical
873: galaxies, and so NLS1s will end as Seyfert 1's with smaller black holes
874: and hosted by spiral galaxies.
875:
876: Our analyses provide hints about the physical connections between
877: different classes of AGNs. It will be important in the future to
878: quantify how AGNs evolve in the multiple-dimensional space of accretion rate,
879: black hole mass, host galaxies etc. With multi-wavelength observations
880: and data from large surveys coming in
881: from both space and ground-based observatories, this task appears to be
882: increasingly achievable.
883:
884: \acknowledgments
885: We would like to thank Drs. X. W. Cao, G. Fazio,
886: J. S. Huang, J. Y. Wei and X. Z. Zheng for advice and helpful
887: discussions. Thanks are also due to Dr A. Laor for advice on
888: how to derive the continuum flux densities for PG QSOs.
889: We also thank an anonymous referee and the editor, Dr. J. Shields, for very
890: constructive comments that improved the paper.
891: This project is supported by the NSF of
892: China No. 10333060, No. 10273012 and TG1999075404.
893: SM acknowledges the financial support of the Chinese Academy of Sciences
894: and the hospitalities of Shanghai Astronomical Observatory in particular
895: Dr. Y. P. Jing during several visits in 2004.
896:
897: \clearpage
898: \begin{references}
899: \reference{}Alexander, D. M. et al. 2004, in The ESO/USM/MPE Workshop
900: on "Multiwavelength Mapping of Galaxy Formation and Evolution", eds. R. Bender \&
901: A. Renzini, in press (astro-ph/0401129)
902: \reference{}Armus, L., et al. 2004, \apjs, 154, 178
903: \reference{}Barthel, P. 2001, NewAR, 45, 591
904: \reference{}Boroson, T. A., \& Green, R. F. 1992, \apjs, 80, 109 (BG92)
905: \reference{}Canalizo, G., \& Stockton, A. 2001, \apj, 555, 719
906: \reference{}Cardiel, N., Elbaz, D., Schiavon, R. P., Willmer, C. N. A., Koo, D. C., Phillips, A. C., \& Gallego, J. 2003, \apj, 584, 76
907: \reference{}Carilli, C. L., et al. 2004, \aj, 128, 997
908: \reference{}Carilli, C. L., Bertoldi, F., Omont, A., Cox, P., McMahon, R. G., \& Isaak, K. G. 2001, \aj, 122, 1679
909: \reference{}Clements, D. L., Sutherland, W. J., Saunders, W., Efstathiou, G. P., McMahon, R. G., Maddox, S., Lawrence, A., \& Rowan-Robinson, M. 1996, \mnras, 279, 459
910: \reference{}Collin, S., Boisson, C., Mouchet, M., Dumont, A.-M., Coup\'e, S., Porquet, D., \& Rokaki, E. 2002, \aap, 388, 771
911: \reference{}Crenshaw, D. M., Kraemer, S. B., Gabel, J. R. 2003, \aj, 126, 1690
912: \reference{}Cui, J., Xia, X. Y., Deng, Z. G., Mao, S., \& Zou, Z. L. 2001, \aj, 122, 63
913: \reference{}Downes, D., \& Solomon, P.M. 1998, \apj, 507, 615
914: \reference{}Elbaz, D., Cesarsky, C. J., Chanial, P., Aussel, H., Franceschini, A., Fadda, D., \& Chary, R. R. 2002, \aap, 384, 848
915: \reference{}Ferrarese, L., \& Merritt, D. 2000, \apjl, 539, L9
916: \reference{}Franceschini, A., et al. 2003, \mnras, 343, 1181
917: \reference{}Genzel, R., et al. 1998, \apj, 498, 579
918: \reference{}Goldader, J. D., Joseph, R. D.; Doyon, R., \& Sanders, D. B. 1995, \apj, 444, 97
919: \reference{}Grupe, D. 2004, \aj, 127, 1799
920: \reference{}Haas, M., M\"uller, S. A. H., Chini, R., Meisenheimer, K., Klaas, U., Lemke, D., Kreysa, E., \& Camenzind, M. 2000, \aap, 354, 453
921: \reference{}Haas, M., et al. 2003, \aap, 402, 87
922: \reference{}Heckman, T. M., Kauffmann, G., Brinchmann, J., Charlot, S., Tremonti, C., \& White, S. D. M. 2004, \apj, 613, 109
923: \reference{}Hughes, D. H., et al. 1998, Nature, 394, 241
924: \reference{}Isobe, T., Feigelson, E. D., \& Nelson, P. I. 1986, \apj, 306, 490
925: \reference{}Kaspi, S., Smith, P. S., Netzer, H., Maoz, D., Jannuzi, B. T., \& Giveon, U. 2000, \apj, 533, 631
926: \reference{}Kauffmann, G., \& Heckman, T. M. 2004, Philosophical Transactions of the Royal Society, in press (astro-ph/0406219)
927: \reference{}Kennicutt, R. C. 1998, \araa, 36, 189
928: \reference{}Kim, D.-C., \& Sanders, D. B. 1998, \apjs, 119, 41
929: \reference{}Kim, D.-C., Veilleux, S., \& Sanders, D. B. 1998, \apj, 508, 627
930: \reference{}Komossa, S., Burwitz, V., Hasinger, G., Predehl, P., Kaastra, J. S., \& Ikebe, Y. 2003, \apj, 582, L15
931: \reference{}Kormendy, J., \& Gebhardt, K. 2001, in The 20th Texas Symposium on Relativistic Astrophysics, ed. H. Martel \& J.C. Wheeler, AIP, in press (astro-ph/0105230)
932: \reference{}Laor, A. 1998, \apjl, 505, L83
933: \reference{}Lawrence, A., Rowan-Robinson, M., Leech, K., Jones, D. H. P., \& Wall, J. V. 1989, \mnras, 240, 329
934: \reference{}Lawrence, A., et al. 1999, \mnras, 308, 897
935: \reference{}Lutz, D., Spoon, H. W. W., Rigopoulou, D., Moorwood, A. F. M., \& Genzel, R. 1998, \apjl, 505, L103
936: \reference{}Magorrian, J., et al. 1998, \aj, 115, 2285
937: \reference{}Marconi, A., \& Hunt, L. K. 2003, \apjl, 589, L21
938: \reference{}Marconi, A., Risaliti, G., Gilli, R., Hunt, L. K., Maiolino, R., \& Salvati, M. 2004, \mnras, 351, 169
939: \reference{}Marziani, P., Zamanov, R., Sulentic, J. W., \& Calvani, M. 2003, \mnras, 345, 1133
940: \reference{}McLure, R. J., \& Jarvis, M. J. 2002, \mnras, 337, 109
941: \reference{}Murphy, T. W., Jr., Armus, L., Matthews, K., Soifer, B. T., Mazzarella, J. M., Shupe, D. L., Strauss, M. A., \& Neugebauer, G. 1996, \mnras, 111, 1025
942: \reference{}Murphy, T. W., Jr., Soifer, B. T., Matthews, K., Armus, L., \& Kiger, J. R. 2001, \aj, 121, 97
943: \reference{}Neugebauer, G., Green, R. F., Matthews, K., Schmidt, M., Soifer, B. T., \& Bennett, J. 1987, \apjs, 63, 615
944: \reference{}Osterbrock, D. E. 1989, Astrophysics of Gaseous Nebulae and Active Galactic Nuclei (Mill Valley: University Science)
945: \reference{}Peeters, E., Spoon, H. W. W., \& Tielens, A. G. G. M. 2004, \apj, 613, 986
946: \reference{}Peterson, B. M. 1997, An Introduction to Active Galactic Nuclei (Cambridge: Cambridge University Press)
947: \reference{}Peterson, B. M., \& Wandel, A. 1999, \apjl, 521, L95
948: \reference{}Peterson, B. M., \& Wandel, A. 2000, \apjl, 540, L13
949: \reference{}Rigopoulou, D., Spoon, H. W. W., Genzel, R., Lutz, D., Moorwood, A. F. M., \& Tran, Q. D. 1999, \aj, 118, 2625
950: \reference{}Sanders, D. B., Soifer, B. T., Elias, J. H., Madore, B. F., Matthews, K., Neugebauer, G., \& Scoville, N. Z. 1988, \apj, 325, 74 (1988a)
951: \reference{}Sanders, D. B., Soifer, B. T., Elias, J. H., Neugebauer, G., \& Matthews, K. 1988, \apjl, 328, L35 (1988b)
952: \reference{}Sanders, D. B., Phinney, E. S., Neugebauer, G., Soifer, B. T., \& Matthews, K. 1989, \apj, 347, 29
953: \reference{}Sanders, D. B., \& Mirabel, I. F. 1996, \araa, 34, 749
954: \reference{}Schmidt, M., \& Green, R. F. 1983, \apj, 269, 352
955: \reference{}Sekiguchi, K. 1987, \apj, 316, 145
956: \reference{}Shemmer, O., \& Netzer, H. 2002, \apjl, 567, L19
957: \reference{}Shemmer, O., Netzer, H., Maiolino, R., Oliva, E., Croom, S., Corbett, E., \& di Fabrizio, L. 2004, \apj, 614, 547
958: \reference{}Shuder, J. M. 1981, \apj, 244, 12
959: \reference{}Springel, V., Di Matteo, T., \& Hernquist, L. 2004, \mnras, submitted (astro-ph/0411108)
960: \reference{}Tran, Q. D., et al. 2001, \apj, 552, 527
961: \reference{}Veilleux, S., Kim, D.-C., \& Sanders, D. B. 1999, \apj, 522, 113 (1999a)
962: \reference{}Veilleux, S., Kim, D.-C., \& Sanders, D. B. 2002, \apjs, 143, 315
963: \reference{}Veilleux, S., Sanders, D. B., \& Kim, D.-C. 1997, \apj, 484, 92
964: \reference{}Veilleux, S., Sanders, D. B., \& Kim, D.-C. 1999, \apj, 522, 139 (1999b)
965: \reference{}Veron-Cetty, M.-P., Veron, P., \& Goncalves, A. C. 2001, \aap, 372, 730
966: \reference{}Wang, J. M. 2003, \aj, 125, 2859
967: \reference{}Wang, T. G., \& Lu, Y. J. 2001, \aap, 377, 52
968: \reference{}Wang, W.-H., Cowie, L. L., \& Barger, A. J. 2004, \apj, 613, 655
969: \reference{}Wu, H., Zou, Z. L., Xia, X. Y., \& Deng, Z. G. 1998, A\&AS, 132, 181
970: \reference{}Xia, X. Y., Boller, Th., Deng, Z. G., \& B\"orner, G. 2001, ChJAA, 1, 221
971: \reference{}Xia, X. Y., Xue, S. J., Mao, S., Boller, T., Deng, Z. G., \& Wu, H. 2002, \apj, 564, 196
972: \reference{}Xu, C., Lonsdale, C. J., Shupe, D. L., O'Linger, J., \& Masci, F. 2001, \apj, 562, 179
973: \reference{}Yee, H. K. C. 1980, \apj, 241, 894
974: \reference{}Zheng, X. Z., Xia, X. Y., Mao, S., Wu, H., \& Deng, Z. G. 2002, \aj, 124, 18
975: \reference{}Zheng, Z., Wu, H., Mao, S., Xia, X. Y., Deng, Z. G., \& Zou, Z. L. 1999, \aap, 349, 735
976: \end{references}
977:
978: \begin{deluxetable}{ccccccccc}
979: \tablecolumns{9}
980: %\noalign{\smallskip}
981: %\noalign{\smallskip}
982: \tabletypesize{\footnotesize}
983: \tablewidth{0pt}
984: \tablecaption{Various Physical Parameters}
985: \tablehead{
986: \colhead{Name} &
987: \colhead{Redshift} &
988: \colhead{log$\left (M_{\rm BH} \over M_\odot \right)$} &
989: \colhead{log$\left (L_{\rm IR} \over L_\odot \right)$} &
990: \colhead{log$\left (L_{\rm opt} \over L_\odot \right)$} &
991: \colhead{log$\left (L_{\rm bol} \over L_\odot \right)$} &
992: \colhead{log$\left ({L \over L_{Edd}}\right)$} &
993: \colhead{log$\left (L_{60\mu{\rm m}} \over L_\odot \right)$} &
994: \colhead{log$\left (L_{\rm H \beta}\over L_\odot \right)$} \\
995: \colhead{(1)} & \colhead{(2)} & \colhead{(3)} & \colhead{(4)} &
996: \colhead{(5)} & \colhead{(6)} & \colhead{(7)} & \colhead{(8)} & \colhead{(9)}}
997: \startdata
998: \cutinhead{IR QSOs}
999: F00275$-$2859 & 0.279 & 7.714 & 12.713\tablenotemark{a} & 11.568 & 12.522 & 0.292 & 12.342 & 9.739 \\
1000: F01572$+$0009 & 0.163 & 7.699 & 12.628 & 11.192 & 12.146 & $-$0.069 & 12.326 & 9.123 \\
1001: F02054$+$0835 & 0.345 & 8.234 & 13.121\tablenotemark{a} & 11.819 & 12.774 & 0.024 & 12.466 & 9.635 \\
1002: F02065$+$4705 & 0.132 & 7.183 & 12.215\tablenotemark{a} & 10.912 & 11.867 & 0.168 & 11.804 & 8.789 \\
1003: F04415$+$1215 & 0.089 & 7.085 & 12.272\tablenotemark{a} & 10.396 & 11.350 & $-$0.250 & 11.905\tablenotemark{b} & 8.395 \\
1004: IR06269$-$0543 & 0.117 & 7.501 & 12.497 & 11.299 & 12.253 & 0.237 & 12.158 & 9.514 \\
1005: F07599$+$6508 & 0.148 & 8.323 & 12.538 & 11.637 & 12.591 & $-$0.248 & 12.116 & 9.603 \\
1006: F09427$+$1929 & 0.284 & 7.568 & 12.715\tablenotemark{a} & 10.977 & 11.931 & $-$0.153 & 12.210 & 8.985 \\
1007: F10026$+$4347 & 0.178 & 7.826 & 12.318\tablenotemark{a} & 11.215 & 12.170 & $-$0.172 & 11.810 & 9.066 \\
1008: F11119$+$3257 & 0.189 & 8.195 & 12.663 & 12.089 & 13.043 & 0.332 & 12.322 & 10.090 \\
1009: Z11598$-$0112 & 0.151 & 6.572 & 12.682\tablenotemark{a} & 10.683 & 11.637 & 0.549 & 12.288 & 8.118 \\
1010: F12134$+$5459 & 0.150 & 6.452 & 12.127\tablenotemark{a} & 10.632 & 11.586 & 0.619 & 11.711 & 8.448 \\
1011: F12265$+$0219 & 0.158 & 8.834 & 12.811 & 12.427 & 13.381 & 0.031 & 12.263 & 10.613 \\
1012: F12540$+$5708 & 0.042 & 8.214 & 12.549 & 11.467 & 12.421 & $-$0.308 & 12.236 & 9.303 \\
1013: F13218$+$0552 & 0.205 & 7.150 & 12.728 & 11.113 & 12.067 & 0.401 & 12.270 & \nodata \\
1014: F13342$+$3932 & 0.179 & 7.421 & 12.496\tablenotemark{a} & 11.821 & 12.775 & 0.838 & 12.116 & 9.657 \\
1015: F15069$+$1808 & 0.171 & 7.000 & 12.249\tablenotemark{a} & 10.696 & 11.651 & 0.135 & 11.861 & 8.651 \\
1016: F15462$-$0450 & 0.101 & 6.889 & 12.250\tablenotemark{a} & 10.381 & 11.335 & -0.070 & 11.995 & 8.361 \\
1017: F16136$+$6550 & 0.129 & 8.996 & 12.003\tablenotemark{a} & 11.550 & 12.504 & $-$1.008 & 11.533 & 9.621 \\
1018: F18216$+$6419 & 0.297 & 9.348 & 13.157\tablenotemark{a} & 12.607 & 13.561 & $-$0.303 & 12.659 & 10.802 \\
1019: F20036$-$1547 & 0.193 & 7.675 & 12.670\tablenotemark{a} & 11.566 & 12.521 & 0.330 & 12.359 & 9.242 \\
1020: F20520$-$2329 & 0.206 & 7.693 & 12.555\tablenotemark{a} & 11.516 & 12.471 & 0.262 & 12.110 & 9.256 \\
1021: F21219$-$1757 & 0.113 & 7.545 & 12.145 & 10.952 & 11.906 & -0.154 & 11.661 & 8.890 \\
1022: F22454$-$1744 & 0.117 & 6.716 & 12.124\tablenotemark{a} & 10.818 & 11.772 & 0.541 & 11.563 & 8.632 \\
1023: F23411$+$0228 & 0.091 & 7.003 & 12.084\tablenotemark{a} & 11.150 & 12.104 & 0.585 & 11.790 & \nodata \\
1024: F01348$+$3254 & 0.367 & 8.532 & 13.018\tablenotemark{a} & 11.991 & 12.946 & $-$0.102 & 12.648 & 9.922 \\
1025: IR03335$+$4729 & 0.184 & 8.107 & 12.686\tablenotemark{a} & 11.998 & 12.952 & 0.330 & 12.112 & 10.043 \\
1026: F04505$-$2958 & 0.286 & 7.791 & 12.723 & 11.824 & 12.778 & 0.471 & 12.341 & 9.798 \\
1027: PG0050$+$124 & 0.061 & 7.155 & 11.970 & 11.050 & 12.004 & 0.333 & 11.310 & 9.050 \\
1028: PG1543$+$489 & 0.400 & 7.838 & 12.784 & 11.843 & 12.797 & 0.443 & 12.344 & 10.039 \\
1029: PG1700$+$518 & 0.292 & 8.307 & 12.703 & 12.115 & 13.070 & 0.247 & 12.090 & 10.148 \\
1030: \cutinhead{PG QSOs}
1031: PG0003$+$158 & 0.450 & 9.068 & 12.784\tablenotemark{a} & 12.270 & 13.224 & -0.359 & 12.298\tablenotemark{b} & 10.521 \\
1032: PG0007$+$106 & 0.089 & 8.416 & 11.429\tablenotemark{a} & 11.103 & 12.058 & -0.874 & 10.645 & 9.400 \\
1033: PG0043$+$039 & 0.384 & 8.992 & 12.279\tablenotemark{a} & 11.992 & 12.946 & -0.561 & 12.303\tablenotemark{b} & 10.248 \\
1034: PG0052$+$251 & 0.155 & 8.639 & 11.735 & 11.441 & 12.395 & -0.759 & 11.080\tablenotemark{b} & 9.673 \\
1035: PG0804$+$761 & 0.100 & 8.131 & 11.733 & 11.360 & 12.314 & -0.333 & 10.794 & 9.728 \\
1036: PG0838$+$770 & 0.131 & 7.841 & 11.571 & 11.020 & 11.974 & -0.383 & 10.994 & 9.321 \\
1037: PG0923$+$201 & 0.190 & 8.936 & 12.335\tablenotemark{a} & 11.386 & 12.340 & -1.112 & 11.684 & 9.880 \\
1038: PG1100$+$772 & 0.313 & 9.133 & 12.040 & 12.011 & 12.965 & -0.684 & 11.396 & 10.258 \\
1039: PG1114$+$445 & 0.144 & 8.309 & 11.893 & 11.084 & 12.038 & -0.787 & 11.236\tablenotemark{b} & 9.376 \\
1040: PG1116$+$215 & 0.177 & 8.319 & 12.206\tablenotemark{a} & 11.739 & 12.693 & -0.142 & 11.398\tablenotemark{b} & 10.274 \\
1041: PG1149$-$110 & 0.049 & 7.481 & 11.061 & 10.299 & 11.253 & -0.744 & 10.435 & 8.663 \\
1042: PG1216$+$069 & 0.334 & 9.008 & 12.487\tablenotemark{a} & 12.049 & 13.003 & -0.521 & 11.804\tablenotemark{b} & 10.281 \\
1043: PG1229$+$204 & 0.064 & 7.841 & 11.139 & 10.756 & 11.710 & -0.647 & 10.492 & 9.074 \\
1044: PG1259$+$593 & 0.472 & 8.778 & 12.778\tablenotemark{a} & 12.279 & 13.233 & -0.061 & 12.187\tablenotemark{b} & 10.357 \\
1045: PG1302$-$102 & 0.286 & 8.712 & 12.353 & 12.165 & 13.120 & -0.108 & 12.063 & 9.905 \\
1046: PG1307$+$085 & 0.155 & 8.640 & 11.660 & 11.411 & 12.365 & -0.791 & 11.257 & 9.811 \\
1047: PG1309$+$355 & 0.184 & 8.138 & 11.957\tablenotemark{a} & 11.433 & 12.387 & -0.267 & 11.305\tablenotemark{b} & 9.433 \\
1048: PG1322$+$659 & 0.168 & 7.985 & 11.611 & 11.256 & 12.210 & -0.291 & 10.962 & 9.435 \\
1049: PG1351$+$236 & 0.055 & 8.209 & 11.163\tablenotemark{a} & 10.411 & 11.365 & -1.360 & 10.450 & 8.065 \\
1050: PG1351$+$640 & 0.087 & 8.567 & 11.789 & 11.203 & 12.157 & -0.926 & 11.248 & 9.259 \\
1051: PG1352$+$183 & 0.158 & 8.158 & 11.754\tablenotemark{a} & 11.176 & 12.130 & -0.543 & 11.244 & 9.592 \\
1052: PG1354$+$213 & 0.300 & 8.362 & 12.035\tablenotemark{a} & 11.312 & 12.266 & -0.612 & 11.997\tablenotemark{b} & 9.368 \\
1053: PG1411$+$442 & 0.089 & 7.757 & 11.533 & 10.945 & 11.899 & -0.374 & 10.579 & 9.263 \\
1054: PG1415$+$451 & 0.114 & 7.709 & 11.442 & 10.894 & 11.848 & -0.377 & 10.683 & 8.950 \\
1055: PG1416$-$129 & 0.129 & 8.752 & 11.606\tablenotemark{a} & 11.397 & 12.351 & -0.917 & 10.868\tablenotemark{b} & 9.930 \\
1056: PG1425$+$267 & 0.366 & 9.533 & 12.379\tablenotemark{a} & 12.062 & 13.017 & -1.032 & \nodata & 10.323 \\
1057: PG1426$+$015 & 0.086 & 8.723 & 11.642 & 11.194 & 12.148 & -1.091 & 10.753 & 9.431 \\
1058: PG1427$+$480 & 0.221 & 7.836 & 11.726 & 11.145 & 12.099 & -0.253 & 11.187 & 9.574 \\
1059: PG1435$-$067 & 0.129 & 8.074 & 11.680 & 11.215 & 12.169 & -0.420 & 11.240 & 9.660 \\
1060: PG1444$+$407 & 0.267 & 8.115 & 12.237 & 11.636 & 12.590 & -0.040 & 11.556 & 9.786 \\
1061: PG1501$+$106 & 0.036 & 8.200 & 11.025 & 10.649 & 11.603 & -1.112 & 10.468 & 9.084 \\
1062: PG1512$+$370 & 0.371 & 9.141 & 12.351 & 11.881 & 12.835 & -0.822 & 11.603 & 10.260 \\
1063: PG1519$+$226 & 0.137 & 7.631 & 11.723\tablenotemark{a} & 11.001 & 11.956 & -0.192 & 10.963\tablenotemark{b} & 9.315 \\
1064: PG1534$+$580 & 0.030 & 7.825 & 10.493\tablenotemark{a} & 10.069 & 11.023 & -1.318 & 9.663 & 8.348 \\
1065: PG1545$+$210 & 0.266 & 9.098 & 11.915 & 11.764 & 12.718 & -0.895 & 11.109 & 10.039 \\
1066: PG1612$+$261 & 0.131 & 7.752 & 11.780\tablenotemark{a} & 11.019 & 11.973 & -0.295 & 11.053 & 9.557 \\
1067: PG1626$+$554 & 0.133 & 8.207 & 11.393 & 10.942 & 11.897 & -0.826 & 10.979\tablenotemark{b} & 9.390 \\
1068: PG1704$+$608 & 0.371 & 9.260 & 12.648 & 12.129 & 13.083 & -0.693 & 12.094 & 9.869 \\
1069: PG2112$+$059 & 0.466 & 8.851 & 12.817\tablenotemark{a} & 12.484 & 13.438 & 0.072 & 12.155\tablenotemark{b} & 10.856 \\
1070: PG2130$+$099 & 0.061 & 7.597 & 11.521 & 10.876 & 11.830 & -0.282 & 10.748 & 9.206 \\
1071: PG2308$+$098 & 0.432 & 9.407 & 12.592\tablenotemark{a} & 12.092 & 13.046 & -0.876 & 12.202\tablenotemark{b} & 10.339 \\
1072: \cutinhead{NLS1s}
1073: TonS180 & 0.062 & 7.091 & 11.259\tablenotemark{a} & 11.127 & 12.081 & 0.474 & 10.530 & 9.082 \\
1074: Mrk359 & 0.017 & 6.329 & 10.416 & 10.144 & 11.098 & 0.253 & 9.982 & 7.692 \\
1075: Mrk1044 & 0.016 & 6.339 & 10.082 & 9.997 & 10.951 & 0.096 & 9.510 & 8.089 \\
1076: IR04312$+$4008 & 0.020 & 6.673 & 10.874\tablenotemark{a} & 10.772 & 11.726 & 0.538 & 10.500 & 8.269 \\
1077: IR04576$+$0912 & 0.037 & 6.581 & 11.178 & 10.136 & 11.090 & -0.007 & 10.835 & 7.542 \\
1078: IR05262$+$4432 & 0.032 & 7.257 & 11.168\tablenotemark{a} & 11.944 & 12.898 & 1.125 & 10.760 & 9.598 \\
1079: Mrk382 & 0.034 & 6.712 & 10.755\tablenotemark{a} & 10.270 & 11.224 & -0.004 & 9.875 & 7.863 \\
1080: Mrk124 & 0.056 & 7.270 & 11.288 & 10.667 & 11.622 & -0.164 & 10.824 & 8.593 \\
1081: Mrk1239 & 0.019 & 6.482 & 10.895\tablenotemark{a} & 10.141 & 11.096 & 0.098 & 10.151 & 8.326 \\
1082: IR09571$+$8435 & 0.092 & 6.801 & 11.573 & 10.525 & 11.479 & 0.163 & 11.029 & 8.232 \\
1083: PG1011-040 & 0.058 & 7.094 & 10.963\tablenotemark{a} & 10.714 & 11.668 & 0.058 & 10.233 & 8.619 \\
1084: PG1016$+$336 & 0.024 & 6.538 & 10.265\tablenotemark{a} & 9.676 & 10.631 & -0.423 & 9.519 & 7.781 \\
1085: Mrk142 & 0.045 & 6.769 & 10.825 & 10.266 & 11.221 & -0.064 & 9.977 & 8.337 \\
1086: KUG1031$+$398 & 0.042 & 6.451 & 10.917\tablenotemark{a} & 10.289 & 11.243 & 0.277 & 10.271 & 7.812 \\
1087: Mrk42 & 0.024 & 6.135 & 10.367\tablenotemark{a} & 9.882 & 10.836 & 0.186 & 9.734 & 7.731 \\
1088: NGC4051 & 0.002 & 5.623 & 9.409 & 8.675 & 9.629 & -0.510 & 8.913 & 6.666 \\
1089: Mrk766 & 0.012 & 6.749 & 10.611 & 9.987 & 10.941 & -0.324 & 10.229 & 8.020 \\
1090: NGC4748 & 0.014 & 6.732 & 10.289 & 10.016 & 10.970 & -0.277 & 9.823 & 8.121 \\
1091: Mrk783 & 0.067 & 7.217 & 11.284\tablenotemark{a} & 10.732 & 11.687 & -0.046 & 10.643 & 8.648 \\
1092: Mrk684 & 0.046 & 6.902 & 10.973 & 10.734 & 11.688 & 0.270 & 10.450 & 8.628 \\
1093: PG1448$+$273 & 0.065 & 6.964 & 11.007\tablenotemark{a} & 10.965 & 11.920 & 0.440 & 10.193 & 8.814 \\
1094: Mrk486 & 0.038 & 7.107 & 10.669\tablenotemark{a} & 10.530 & 11.484 & -0.138 & 9.742\tablenotemark{b} & 8.912 \\
1095: Mrk493 & 0.031 & 6.213 & 10.738 & 10.232 & 11.187 & 0.458 & 10.301 & 8.187 \\
1096: B31702$+$457 & 0.060 & 6.822 & 11.546 & 10.837 & 11.791 & 0.454 & 11.116 & 8.408 \\
1097: Mrk507 & 0.053 & 7.128 & 11.056 & 10.666 & 11.620 & -0.024 & 10.676 & 7.737 \\
1098: HS1817$+$5342 & 0.080 & 7.447 & 11.430\tablenotemark{a} & 11.144 & 12.098 & 0.135 & 10.644 & 9.334 \\
1099: Mrk896 & 0.027 & 6.668 & 10.567 & 10.368 & 11.323 & 0.139 & 10.048 & 8.166 \\
1100: Akn564 & 0.025 & 6.532 & 10.835\tablenotemark{a} & 10.533 & 11.488 & 0.440 & 10.187 & 8.507 \\
1101: PG0003$+$199 & 0.025 & 6.995 & 10.809\tablenotemark{a} & 10.389 & 11.343 & -0.167 & 9.805 & 8.659 \\
1102: PG0026$+$129 & 0.142 & 7.708 & 11.493\tablenotemark{a} & 11.379 & 12.333 & 0.109 & 11.206\tablenotemark{b} & 9.484 \\
1103: PG0923$+$129 & 0.029 & 7.029 & 10.708 & 10.170 & 11.124 & -0.421 & 10.172 & 8.343 \\
1104: PG1001$+$054 & 0.161 & 7.515 & 11.852 & 11.158 & 12.112 & 0.081 & 11.113 & 9.400 \\
1105: PG1119$+$120 & 0.049 & 7.158 & 11.149 & 10.509 & 11.463 & -0.211 & 10.536 & 8.492 \\
1106: PG1211$+$143 & 0.085 & 7.730 & 11.720 & 11.414 & 12.369 & 0.123 & 11.084 & 9.631 \\
1107: PG1244$+$026 & 0.048 & 6.258 & 10.945 & 10.151 & 11.105 & 0.331 & 10.416\tablenotemark{b} & 8.056 \\
1108: PG1402$+$261 & 0.164 & 7.701 & 11.961 & 11.331 & 12.285 & 0.068 & 11.305 & 9.504 \\
1109: PG1404$+$226 & 0.098 & 6.638 & 10.949 & 10.690 & 11.644 & 0.491 & 10.659 & 8.715 \\
1110: PG1440$+$356 & 0.077 & 7.231 & 11.605 & 10.952 & 11.907 & 0.159 & 11.055 & 9.051 \\
1111: \enddata
1112: \tablecomments{Col:(1) name (the prefix IR denotes the IRAS name). Col:(2) redshift. Col:(3) black hole mass. Col:(4) infrared luminosity. Col:(5) monochromatic luminosity at 5100\AA, $\lambda L_\lambda$(5100\AA). Col:(6) bolometric luminosity of AGN (9$\lambda L_\lambda$(5100\AA)). Col:(7) Eddington ratio. Col:(8) monochromatic luminosity
1113: at 60$\mu$m ($\nu L_\nu$). Col:(9) H$\beta$ luminosity.}
1114: \tablenotetext{a}{The flux density in at least one IRAS band is an
1115: upper limit for objects whose
1116: infrared properties were taken from IRAS; the infrared luminosity
1117: in at least one band is an upper limit for objects from Table 2 of
1118: Haas et al. (2003).}
1119: \tablenotetext{b}{The $60\mu {\rm m}$ flux density is an upper limit.}
1120: \end{deluxetable}
1121:
1122: \clearpage
1123:
1124: \begin{deluxetable}{ccccccc}
1125: \tablecolumns{7}
1126: %\noalign{\smallskip}
1127: \tabletypesize{\footnotesize}
1128: \tablewidth{0pt}
1129: \rotate
1130: \tablecaption{Significance of Various Correlations}
1131: \tablehead{
1132: \colhead{Relation} &
1133: \colhead{Sample} &
1134: \colhead{Num\tablenotemark{a}} &
1135: \colhead{r$_s$} &
1136: \colhead{Sig(\%)} &
1137: \colhead{a} &
1138: \colhead{b} \\
1139: \colhead{(1)} & \colhead{(2)} & \colhead{(3)} & \colhead{(4)} &
1140: \colhead{(5)} & \colhead{(6)} & \colhead{(7)}}
1141: \startdata
1142: $L_{\rm H\beta}$ vs. $L_{\rm op}$ & IR + PG + NLS1s & 107(0) & 0.945 & $>99.99$ &
1143: 1.097$\pm$0.039\phn & -3.026$\pm$0.432\phn \\
1144: ~~ & IR QSOs & 29(0) & 0.984 & $>99.99$ &
1145: 1.115$\pm$0.051\phn & -3.365$\pm$0.580\phn \\
1146: ~~ & PG QSOs + NLS1s & 78(0) & 0.950 & $>99.99$ &
1147: 1.140$\pm$0.048\phn & -3.447$\pm$0.531\phn \\
1148: $L_{\rm IR}$ vs. $L_{\rm op}$ & IR QSOs & 31(19) & 0.534 & 99.66 &
1149: 0.505$\pm$0.118\phn & 6.547$\pm$1.364\phn \\
1150: ~~ & PG QSOs + NLS1s & 78(34) & 0.590 & $>99.99$ &
1151: 0.833$\pm$0.063\phn & 2.148$\pm$0.693\phn \\
1152: $L_{\rm 12um}$ vs. $L_{\rm op}$ & IR QSOs & 31(15) & 0.583 & 99.86 &
1153: 0.555$\pm$0.126\phn & 5.426$\pm$1.451\phn \\
1154: ~~ & PG QSOs + NLS1s & 77(18) & 0.802 & $>99.99$ &
1155: 0.910$\pm$0.055\phn & 0.921$\pm$0.606\phn \\
1156: $L_{\rm 25um}$ vs. $L_{\rm op}$ & IR QSOs & 31(6) & 0.737 & 99.99 &
1157: 0.510$\pm$0.070\phn & 6.054$\pm$0.794\phn \\
1158: ~~ & PG QSOs + NLS1s & 70(14) & 0.794 & $>99.99$ &
1159: 0.951$\pm$0.067\phn & 0.440$\pm$0.728\phn \\
1160: $L_{\rm 60um}$ vs. $L_{\rm op}$ & IR QSOs & 31(1) & 0.585 & 99.86 &
1161: 0.352$\pm$0.085\phn & 8.062$\pm$0.975\phn \\
1162: ~~ & PG QSOs + NLS1s & 77(17) & 0.657 & $>99.99$ &
1163: 0.794$\pm$0.062\phn & 2.016$\pm$0.678\phn \\
1164: $L_{\rm 100um}$ vs. $L_{\rm op}$ & IR QSOs & 31(5) & 0.491 & 99.29 &
1165: 0.296$\pm$0.090\phn & 8.547$\pm$1.028\phn \\
1166: ~~ & PG QSOs + NLS1s & 76(30) & 0.478 & $>99.99$ &
1167: 0.591$\pm$0.081\phn & 4.070$\pm$0.884\phn \\
1168: $SFR$ vs. $\dot{M}$ & IR QSOs & 31(1) & 0.491 & 99.29 &
1169: 0.291$\pm$0.100\phn & 2.771$\pm$0.059\phn \\
1170: \\
1171: $SFR/\dot{M}$ vs. $M_{\rm bh}$ & IR QSOs & 31(1) & -0.670 & 99.98 &
1172: -0.518$\pm$0.091\phn & 6.615$\pm$0.708\phn \\
1173: \enddata
1174: \tablecomments{Col:(1) relations. Col:(2) sample. Col:(3) number of sources.
1175: Col:(4) S-R coefficient. Col:(5) significance level. Col: (6) and (7) are the
1176: coefficients of linear regressions for various correlations obtained using the
1177: EM algorithm (ASURV, Isobe, Feigelson \& Nelson 1986): $\log Y=a\log X+b$.}
1178: \tablenotetext{a}{The number in parenthesis denotes the number of sources whose dependent
1179: variables are upper limits involved in the survival analysis.}
1180: \end{deluxetable}
1181:
1182: \clearpage
1183:
1184: \begin{figure}
1185: \figurenum{1}
1186: \epsscale{}
1187: \plotone{f1.eps}
1188: \caption{(a) H$\beta$ luminosity (b) $L_{\rm IR}$ vs. $\lambda L_\lambda$(5100\AA).
1189: The open circles represent IR QSOs,
1190: while the filled circles and triangles represent PG QSOs and NLS1s,
1191: respectively. The open circles enclosed by open squares are the three IR
1192: QSOs that are also PG QSOs.
1193: In panel (a), the solid line represents the linear regression for all IR QSOs, PG QSOs and
1194: NLS1s excluding NGC4051. Note that in all our statistical analysis, NGC4051 is
1195: excluded because it is far from others in terms of all physical parameters. We will not
1196: mention this in later figures.
1197: In panel (b), the solid line represents
1198: the linear regression for all PG QSOs and NLS1s.
1199: For clarity, upper limits of $L_{\rm IR}$ are
1200: not labelled (see Table 1), but survival analyses were performed for
1201: these upper limits.}
1202: \label{lhblirlop.eps}
1203: \end{figure}
1204:
1205: \clearpage
1206:
1207: \begin{figure}
1208: \figurenum{2}
1209: \epsscale{}
1210: \plotone{f2.eps}
1211: \caption{The monochromatic luminosities at $12\mu{\rm m}$, 25$\mu{\rm
1212: m}$, $60\mu{\rm m}$ and $100\mu{\rm m}$ vs. $\lambda L_\lambda$(5100\AA). The
1213: symbols are the same as in Fig.~\ref{lhblirlop.eps} and the arrows denote upper limits. The
1214: solid line represents the linear regression for all the PG
1215: QSOs with available infrared flux densities and NLS1s. The
1216: accretion rate is labelled at the top abscissa. In panel (c) the best-fit
1217: line for IR QSOs is plotted as the dashed line.}
1218: \label{l4band.lop.eps}
1219: \end{figure}
1220:
1221: \clearpage
1222:
1223: \begin{figure}
1224: \figurenum{3a}
1225: \epsscale{}
1226: \plotone{f3a.eps}
1227: \caption{Left: Histograms of $L_{\rm IR}/L_{\rm bol}$.
1228: Right: Histograms of $L_{\rm 60\mu m}/L_{\rm bol}$
1229: for the IR QSOs (top), PG QSOs (middle) and NLS1s (bottom). The median
1230: values are labelled in the panels.
1231: }
1232: \label{his2panel1.eps}
1233: \end{figure}
1234:
1235: \clearpage
1236:
1237: \begin{figure}
1238: \figurenum{3b}
1239: \epsscale{}
1240: \plotone{f3b.eps}
1241: \caption{Left: Histograms of the Eddington accretion
1242: ratio $L_{\rm bol}/L_{\rm Edd}$.
1243: Right: Histograms of the black hole mass
1244: for the IR QSOs (top), PG QSOs (middle) and NLS1s (bottom). The median
1245: values are labelled in the panels.
1246: }
1247: \label{his2panel2.eps}
1248: \end{figure}
1249:
1250:
1251: \clearpage
1252:
1253: \begin{figure}
1254: \figurenum{4}
1255: \epsscale{}
1256: \plotone{f4.eps}
1257: \caption{The infrared
1258: spectral index $\alpha (60,25)$ vs. the infrared excess,
1259: $L_{\rm IR}/L_{\rm bol}$.
1260: The symbols are the same as in Fig.~\ref{lhblirlop.eps}.}
1261: \label{alpha.lirlbol.eps}
1262: \end{figure}
1263:
1264:
1265: \clearpage
1266:
1267:
1268: \begin{figure}
1269: \figurenum{5}
1270: \epsscale{}
1271: \plotone{f5.eps}
1272: \caption{The ratio of the star
1273: formation rate and the accretion rate of black holes vs.
1274: the black hole mass for IR QSOs. The solid line is the best regression
1275: line (see eq. \ref{eq:sfrm}).}
1276: \label{ir.sbsfrmdot.mbh.eps}
1277: \end{figure}
1278:
1279:
1280: \end{document}
1281:
1282: % LocalWords: QSOs NLS Hao Xia Datun Jodrell Macclesfield DL Sceince AGNs AGN
1283: % LocalWords: starbursts Eddington starburst ultraluminous ULIGs circum Genzel
1284: % LocalWords: Veilleux Veilluex Goldader Rigopoulou et al Zheng Canalizo Cui
1285: % LocalWords: Boroson Marziani Grupe XMM NGC Mrk Komossa Franceschini QSO BG
1286: % LocalWords: Downes Carilli comoving Elbaz summarize Mpc QDOT Jy ROSAT PSCz
1287: % LocalWords: Hass Catalog Veron Cetty BH BLR virialized analyses Wandel Kaspi
1288: % LocalWords: McLure lt FWHM dex Neugebauer NIR eq Bolometric bolometric bol
1289: % LocalWords: Cardiel Kennicutt Eqs Spearman ASURV Feigelson Kauffmann Heckman
1290: % LocalWords: Peeters Indices Shemmer Netzer neighboring millimeter color PAH
1291: % LocalWords: Spitzer Armus accreting SFR behavior Ferrarese Kormendy Gebhardt
1292: % LocalWords: Magorrian Laor SDSS emphasize QSO's Springel spheroidals AGN's
1293: % LocalWords: Eddingtion Crenshaw Drs Cao Fazio TG Jing astro ph NewAR Boisson
1294: % LocalWords: Mouchet Coupé Porquet Rokaki Gabel Zou Chartas Garmire Sambruna
1295: % LocalWords: Ménard Brinchmann Tremonti Brinkmann Kawaguchi Aoki Ohta Burwitz
1296: % LocalWords: Hasinger Predehl Kaastra Ikebe laor Tielens Hernquist Shude cD
1297: % LocalWords: MNRAS Sulentic Zwitter Dultzin Hacyan Tacconi Goncalves IPAC Xu
1298: % LocalWords: CHJAA ccccccccc TonS KUG Akn cccccc bh outlier Isobe Yee Shuder
1299: % LocalWords: Osterbrock ionization Sekiguchi analyzed emphasized reradiated
1300: % LocalWords: ary ESO Multiwavelength Renzini Schiavon Willmer Gallego Omont
1301: % LocalWords: Bertoldi Efstathiou Cesarsky Chanial Aussel Fadda Doyon uller di
1302: % LocalWords: Chini Meisenheimer Klaas Kreysa Camenzind Maoz Jannuzi Giveon
1303: % LocalWords: AIP Moorwood Risaliti Maiolino Salvati Zamanov Calvani Soifer
1304: % LocalWords: Mazzarella Shupe Kiger Madore Scoville Phinney Oliva Croom Cowie
1305: % LocalWords: Fabrizio Barger Boller orner ChJAA Xue Lonsdale O'Linger Masci
1306: % LocalWords: ccccccc Num eps ULIG NSL
1307: