1: \documentclass[12pt,preprint]{aastex}

2: %\documentclass[manuscript]{aastex}

3: \newcommand{\vdag}{(v)^\dagger}

4: \newcommand{\myemail}{rwang@nrao.edu}

5: 

6: \shorttitle{Warm Dust Emission in z$\sim$6 Quasars}

7: \shortauthors{Wang et al.}

8: 

9: \begin{document}

10: 

11: \title{Thermal Emission from Warm Dust in the Most Distant Quasars}

12: 

13: \author{Ran Wang\altaffilmark{1,2},

14: Chris L. Carilli\altaffilmark{2},

15: Jeff Wagg\altaffilmark{2}

16: Frank Bertoldi\altaffilmark{3},

17: Fabian Walter\altaffilmark{4},

18: Karl M. Menten\altaffilmark{5},

19: Alain Omont\altaffilmark{6},

20: Pierre Cox\altaffilmark{7},

21: Michael A. Strauss\altaffilmark{8},

22: Xiaohui Fan\altaffilmark{9},

23: Linhua Jiang\altaffilmark{9}

24: Donald P. Schneider\altaffilmark{10}}

25: \altaffiltext{1}{Department of Astronomy, Peking University, Beijing 100871, China}

26: \altaffiltext{2}{National Radio Astronomy Observatory, PO Box 0, Socorro, NM, USA 87801}

27: \altaffiltext{3}{Argelander-Institut f$\rm \ddot u$r Astronomie, University of Bonn, Auf dem H$\rm \ddot u$gel 71, 53121 Bonn, Germany}

28: \altaffiltext{4}{Max-Planck-Institute for Astronomy, K$\rm \ddot o$nigsstuhl 17, 69117 Heidelberg, Germany}

29: \altaffiltext{5}{Max-Planck-Institut f$\rm \ddot u$r Radioastronomie, Auf dem H$\rm \ddot u$gel 71, 53121 Bonn, Germany}

30: \altaffiltext{6}{Institut d'Astrophysique de Paris, CNRS and Universite Pierre et Marie Curie, Paris, France}

31: \altaffiltext{7}{Institute de Radioastronomie Millimetrique, St. Martin d'Heres, F-38406, France}

32: \altaffiltext{8}{Department of Astrophysical Sciences, Princeton University, Princeton, NJ, USA, 08544}

33: \altaffiltext{9}{Steward Observatory, The University of Arizona, Tucson, AZ 85721}

34: \altaffiltext{10}{Department of Astronomy and Astrophysics, The Pennsylvania State University, University Park, PA 16802}

35: \begin{abstract}

36: 

37: We report new continuum observations of fourteen z$\sim$6 

38: quasars at 250 GHz and fourteen quasars at 1.4 GHz. We  

39: summarize all recent millimeter and radio observations of 

40: the sample of the thirty-three quasars known with $\rm 5.71\le z\le 6.43$, 

41: and present a study of the rest frame far-infrared (FIR) 

42: properties of this sample. These quasars were observed with 

43: the Max Plank Millimeter Bolometer Array (MAMBO) at 250 GHz 

44: with mJy sensitivity, and $\rm 30\%$ of them were detected. 

45: We also recover the average 250 GHz flux density of the MAMBO 

46: undetected sources at $\rm 4\sigma$, by stacking the on-source 

47: measurements. The derived mean radio-to-UV spectral energy distributions (SEDs) 

48: of the full sample and the 250 GHz non-detections show no significant 

49: difference from that of lower-redshift 

50: optical quasars. Obvious FIR excesses are seen in 

51: the individual SEDs of the strong 250 GHz detections, with FIR-to-radio 

52: emission ratios consistent with that of typical star forming 

53: galaxies. Most 250 GHz-detected sources follow the 

54: $\rm L_{FIR}$--$\rm L_{bol}$ relationship derived from a sample 

55: of local IR luminous quasars ($\rm L_{IR}>10^{12}L_{\odot}$), 

56: while the average $\rm L_{FIR}/L_{bol}$ ratio of the non-detections 

57: is consistent with that of the optically-selected PG quasars. 

58: The MAMBO detections also tend to have weaker $\rm Ly\alpha$ 

59: emission than the non-detected sources. We discuss possible 

60: FIR dust heating sources, and critically assess the possibility 

61: of active star formation in the host galaxies of the z$\sim$6 

62: quasars. The average star formation rate of the MAMBO non-detections is 

63: likely to be less than a few hundred $\rm M_{\odot}\,yr^{-1}$, 

64: but in the strong detections, the host galaxy star formation is probably at a rate of 

65: $\rm \gtrsim10^{3}\,M_{\odot}\,yr^{-1}$, which dominates the FIR 

66: dust heating.

67: 

68: \end{abstract}

69: 

70: \keywords{galaxies: quasars -- galaxies: active -- submillimeter

71: -- galaxies: starburst -- galaxies: high-redshift}

72: 

73: \section{Introduction}

74: 

75: More than thirty quasars at z$\sim$6 have been discovered 

76: (e.g., Fan et al. 2000, 2001a, 2003, 2004, 2006a; Jiang

77: et al. 2007a; Willott et al. 2007). These objects are

78: the earliest massive black hole accretion

79: systems known (Jiang et al. 2006; 2007b), seen at an epoch

80: close to the end of cosmic reionization (Fan et al. 2006b).

81: They provide us with unique information on both the growth

82: of supermassive black holes (SMBHs) and the formation

83: of massive galaxies when the age of the universe was 

84: $\rm \lesssim1\,Gyr$.

85: %The X-ray to near-infrared observations of these objects

86: %indicate that the central supermassive black

87: %holes (SMBHs) have masses $\rm \gtrsim10^{9}\,M_{\odot}$

88: %and are in a rapid accretion stage, with Eddington

89: %ratios close to unity (Jiang et al. 2006).

90: 

91: There are fundamental relationships between SMBH mass and bulge stellar 

92: mass/velocity dispersion in the local universe (e.g., Tremaine et

93: al. 2002; Marconi \& Hunt 2003; Hopkins et al. 2007), indicating that

94: the formation of SMBHs and their spheroidal hosts are

95: coupled. 

96: %Studies of galaxy evolution suggest such

97: %SMBH-spheroidal systems evolved from galaxy mergers,

98: %namely, merger driven gas inflows fuel the rapid growth 

99: %of both the stellar bulge and the central black hole until 

100: %feedback from the active galactic nucleus (AGN) turns star 

101: %formation off (e.g., Hopkins et al. 2006; 2007b). 

102: Active galactic nuclei (AGNs) buried in dusty starburst environments have been discovered in  

103: samples of low redshift ultraluminous infrared galaxies (ULIRGs), 

104: which are believed to be a transition phase between starburst 

105: mergers and typical optically bright AGNs (Sanders et al. 1988; 

106: Wu et al. 1998; Zheng et al. 2002). 

107: Studies of large samples of galaxies and AGNs 

108: also provide clues about the 'downsizing' process in galaxy and SMBH formation, 

109: whereby the formation of the most massive systems occur at early epochs

110: (Cowie et al. 1996; Heckman et al. 2004; Kauffmann \& Heckman 2005).

111: %Thus we may expect that bulges are built via active star 

112: %formation at the same time that SMBHs accrete and grow rapidly in 

113: %the massive quasar systems at high redshift. 

114: 

115: Examples of massive quasars hosted by interacting systems or 

116: bright submillimeter galaxies have been found at redshifts 

117: greater than 4, such as BR 1202-0725 at z=4.7 and BRI1335-0417

118: at z=4.4 (Omont et al. 1996a, 1996b; Benford et al.

119: 1999; Beelen et al. 2006). These objects are all characterized by

120: strong molecular CO line emission (Guilloteau et al. 1997, 1999; Carilli

121: et al. 2002; Solomon \& Vanden Bout 2005; Riechers et al. 2006) and

122: FIR (Guilloteau et al. 1999) and radio (Momjian et al.

123: 2005, 2007) emission originating from a starburst, with implied star 

124: formation rates of a few thousand $\rm M_{\odot}\,yr^{-1}$. 

125: The results suggest that the galaxies are at an earlier evolution 

126: stage than are typical optically bright quasars in which the central 

127: AGN dominates the emission from X-ray to radio.

128: 

129: These studies have been extended to z$\sim$6 quasar sample with

130: sensitive submillimeter, millimeter, and radio telescopes (Priddey

131: et al. 2003b; Bertoldi et al. 2003; Petric et al. 2003; Carilli et al. 2004; Wang et al.

132: 2007). 

133: %Understanding dust and gas properties of the quasar hosts at

134: %z$\sim$6 provides key leverage into the understanding of the evolution of

135: %the earliest galaxy generations at a time when the age of the

136: %universe is $\lesssim$1 Gyr. 

137: About 1/3 of these z$\sim$6 quasars

138: were detected at millimeter wavelengths, 

139: at mJy sensitivity (Priddey et al. 2003b; Wang et al. 2007). The 

140: millimeter detections imply FIR luminosities of $\rm 10^{12}\sim10^{13}\,L_{\odot}$

141: and dust masses of $\rm \gtrsim10^{8}\,M_{\odot}$ in the quasar host galaxies

142: (Bertoldi et al. 2003a, Beelen et al. 2006). Such huge dust masses

143: require rapid metal and dust enrichment via a starburst in the 

144: early evolution of these galaxies (Bertoldi et al. 2003a; 

145: Walter et al. 2003; Venkatesan et al. 2006; Li et al. 2007b; Dwek et al. 2007).

146: 

147: The heating sources of the FIR-emitting dust in the quasar

148: systems at z$\sim$6 have become a key question. The reprocessed emission from 

149: star formation-heated dust can provide a direct estimate   

150: of the star formation rate, thus constraining the bulge building 

151: stage in these quasar hosts. 

152: Observations at submillimeter wavelengths imply that 

153: the FIR emission in the two sources with the strongest millimeter detections, 

154: SDSS J114816.64+525150.3 at z=6.42 (hereafter J1148+5251) 

155: and SDSS J092721.82+200123.7 at z=5.77 (hereafter J0927+2001), 

156: is from dust components with 

157: temperatures of 50 to 60 K (Beelen et al. 2006; Wang et al. 2008a).

158: Large amounts of highly excited molecular CO were also detected 

159: in the host galaxies of these two sources (Bertoldi et al. 2003b; 

160: Walter et al. 2003; Carilli et al. 2007), as well as strong [C {\small II}] 

161: 158$\,\mu$m ISM gas cooling line emission in J1148+5251 (Maiolino et al. 2005).

162: These results suggest that star formation plays an important 

163: role in the heating of FIR-emitting warm dust. The implied star 

164: formation rate is $\rm \gtrsim10^{3}\,M_{\odot}\,yr^{-1}$, which argues 

165: for active bulge building in these two z$\sim$6 quasars. 

166: 

167: We have a long standing program to study the dust and gas emission

168: from the z$\sim$6 quasars (e.g., Wang et al. 2007, 2008a). In this paper, we 

169: report new observations and present an FIR emission study 

170: of the full sample of thirty-three z$\sim$6 quasars discovered to date, 

171: aiming (i) to give a general view of the FIR emission properties of the sample, and

172: (ii) to further constrain the dust heating and star

173: forming activity in the quasar hosts. We describe the full 

174: quasar sample, the new observations, and summarize the current millimeter and

175: radio results in section 2. The analysis of the full sample is given in Sections 3, 4, and 5.

176: We present a discussion of star formation in Section 6,

177: and give a brief summary in Section 7. We adopt a $\rm \Lambda$-model

178: cosmology with $\rm H_{0}=71km\ s^{-1}\ Mpc^{-1}$, $\rm

179: \Omega_{M}=0.27$ and $\rm \Omega_{\Lambda}=0.73$ throughout this

180: paper (Spergel et al. 2007).

181: 

182: \section{Sample and observations}

183: 

184: \subsection{The sample of z$\sim$6 quasars}

185: 

186: Thirty-three quasars have been discovered at z$\sim$6 

187: ($\rm 5.71 \le z \le 6.43$, see Table 1 and 2). Twenty-two 

188: of these objects were selected from $\sim$8000 deg$^{2}$ 

189: of the Sloan Digital Sky Survey (SDSS, York et al. 2000) imaging data 

190: (Fan et al. 2000, 2001a, 2003, 2004, 2006a, 2008, in prep.; 

191: Goto 2006). These objects typically have 

192: rest frame $\rm 1450\AA$ AB magnitudes $\rm m_{1450}<20$.

193: The SDSS imaging survey is using 

194: a dedicated 2.5m telescope and a large format CCD camera 

195: (Gunn et al. 2006) 

196: at the Apache Point Observatory in New Mexico

197: to obtain images in five broad bands ($\rm u$, $\rm g$, $\rm r$, $\rm i$ and $\rm z$,

198: centered at 3551, 4686, 6166, 7480 and 8932 $\rm \AA$, respectively; 

199: Fukugita et al. 1996) 

200: of high Galactic latitude sky in the Northern Galactic Cap.

201: The imaging data are processed by the astrometric 

202: pipeline (Pier et al. 2003) and photometric pipeline (Lupton et al. 2001), 

203: %with a series of pipelines (Lupton et al. 2001, Pier et al. 2003), 

204: resulting in 

205: astrometric calibration errors of $\rm < 0.1''$ rms per coordinate, and

206: photometric calibration to better than 0.03 mag

207: (Hogg et al. 2001, Smith et al. 2002, Ivezi{\' c} et al. 2004, Tucker 

208: et al. 2006).

209: These data have been made available to the public 

210: in a series of data releases (EDR: Stoughton et al. 2002; DR1 

211: -- DR6:

212: Abazajian et al. 2003, 2004, 2005, Adelman-McCarthy et al. 2006, 

213: 2007, 2008).

214: 

215: Five additional z$\sim$6 quasars were discovered in the SDSS

216: Deep Equatorial Stripe ('Stripe 82') by Jiang et al. (2007a). These five

217: sources were selected from deep optical imaging of 

218: 260 deg$^2$. These objects are about

219: one magnitude fainter in the optical (i.e., $\rm 20<m_{1450}<21.5$) than the

220: primary SDSS sample.

221: One source in this sample, J020332.39+001229.3 (hereafter J0203+0012), was also

222: independently discovered in the UKIRT Infrared

223: Deep Sky Survey (Venemans et al. 2007).

224: 

225: Four new quasars at

226: $\rm z>6$ have been published (Willott et al. 2007)

227: from the Canada-France High-z Quasar

228: Survey (CFHQS). These quasars were selected from

229: imaging of 400 deg$^2$; three of them have

230: $\rm m_{1450}>21$. The fourth source, CFHQS J1509-1749,

231: is comparable in brightness to the faintest source in

232: the primary SDSS sample, but the optical spectrum

233: is much steeper than that of the typical quasars,

234: indicating significant dust reddening.

235: 

236: Another two z$\sim$6 quasars were discovered in the NOAO Deep Wide-Field

237: Survey (Jannuzi et al. 2004) using optical, infrared, and radio data 

238: (McGreer et al. 2006; Cool et al. 2006). One of them, J1427+3312 at z=6.12,

239: was selected by McGreer et al. (2006) from 4 deg$^{2}$ 

240: by matching infrared and radio sources to optical counterparts.

241: It is the brightest radio source in the z$\sim$6 quasar sample. The

242: UV to optical spectrum of this object is likely to be reddened by

243: dust, and the rest frame $\rm 1450\AA$ magnitude derived from near-infrared

244: data is $\rm m_{1450}=20.33$\footnote{derived from the absolute 

245: magnitude of $\rm M_{1450}=-26.4$ presented in McGreen et al. (2006)}. 

246: The other quasar was infrared selected from Spitzer mid-infrared 

247: data, and identified in the AGN and Galaxy Evolution Survey 

248: (Cool et al. 2006) with $\rm m_{1450}=20.62$.  

249: 

250: \subsection{New observations}

251: 

252: Twenty-two sources in the z$\sim$6 quasar sample have published 250 GHz 

253: continuum observations with the Max-Planck 

254: Millimeter Bolometer Array (MAMBO, Kreysa et al. 1998) on the IRAM 30-m

255: telescope (Bertoldi et al. 2003a; Petric et al. 2003; Wang et al. 2007; 

256: Willott et al. 2007; Priddey et al. 2008), and nineteen sources in the 

257: sample have published 1.4 GHz continuum observations with the VLA (Petric et al. 2003; 

258: Carilli et al. 2004; Wang et al. 2007; McGreer et al. 2006). 

259: We report our new MAMBO and VLA observations of totally nineteen z$\sim$6 quasars in this paper. 

260: These sources are unobserved or marginally detected in the previous 

261: observations, and we present new data for them at either 250 GHz or 1.4 GHz.

262: We describe the new observations below and list the measurements for the 

263: nineteen quasars in Table 1. We also summarise 

264: the previous observations for other z$\sim$6 quasars in Table 2.

265: 

266: We present new 250 GHz observations for fourteen z$\sim$6 quasars, including 

267: eleven sources that have never been previously observed. 

268: The new observations have been obtained with the MAMBO-II 117-element 

269: array in Winter 2006-2007, we adopted the same observing

270: mode as was used in the previous observations (Wang et al. 2007),

271: i.e., doing chopping photometry at 2 Hz with $\rm 32''$

272: throw in azimuth. We typically spent 1-3 hours on each source to

273: achieve the sub-mJy rms noise level reached in previous observations.

274: The data were reduced with the standard MOPSIC

275: pipeline (Zylka 1998). Three sources, J1044-0125,

276: J1425+3254, and J2054-0005, were detected from

277: the new data at $\rm >4\sigma$ level. The MAMBO  

278: detection of J1044-0125 was recently reported in 

279: Priddey et al. (2008) from an independent observation 

280: with a 250 GHz flux density of $\rm 2.5\pm0.6$. 

281: We observed this source down to a lower rms noise level, 

282: and our result, $\rm 1.8\pm0.4\,mJy$, is a bit lower compared to the value in 

283: Priddey et al. (2008) but consistent within the error limits (see Table 1).  

284: By including the data from the new observations, we also 

285: confirmed the previous $\rm 3\sigma$ detection in 

286: the source J0818+1722 (Wang et al. 2007) with a 

287: lower rms of 0.4 mJy.

288: 

289: New 1.4 GHz data were obtained with the VLA for fourteen 

290: quasars, including thirteen previously unobserved 

291: sources and a published marginally detected source, 

292: J0927+2001 ($\rm 3.2\sigma$, Wang et al. 2007).

293: The new observations were carried out in 2006 and 2007 with the array in the A 

294: or B configurations, down to an rms noise level of $\rm \sim20\,\mu Jy$

295: for most of the sources. There is a $\sim$1 Jy source in the

296: field of the quasar J0303-0019, leading to a high rms

297: of $\rm 66\,\mu Jy$ in the image. All the data

298: were reduced using AIPS.

299: One source, J0203+0012, was detected by our new observation with a 1.4 GHz flux density 

300: of $\rm S_{\nu}=195\pm22\,\mu Jy$, and the previous detection of J0927+2001 

301: was confirmed at $\rm \gtrsim4\sigma$ level. 

302: 

303: There are fairly bright and extended radio sources in the 

304: 1.4 GHz field images of the CFHQS quasars J0033-0125 and J1641+3755 (Figure 1), though these two quasars are not detected. 

305: One double-lobe radio source is found about $\rm9'$ away from the optical position 

306: of the  quasar J0033-0125, with integrated 1.4 GHz flux density of $\rm \sim23\,mJy$. This radio source was also 

307: detected in the NVSS survey (Condon et al. 1998) and FIRST survey (White et al. 1997), and 

308: there is no published deep optical data for this area yet. 

309: Five radio sources with integrated flux densities $\rm >10\,mJy$ are seen 

310: in the field of the source J1641+3755, including extended 

311: double radio sources, within 10$\rm '$ from the optical quasar position. 

312: According to the FIRST 1.4 GHz source counts (White et al. 1997),

313: about 1.6 source with $\rm f_{1.4GHz}>10\,mJy$ is expected within this amount of sky area.

314: The brightest radio source in the field of J1641+3755 was 

315: identified as a radio-loud AGN with a spectroscopic redshift of z=0.162 (Best et al. 2005).

316: Further optical imaging and spectrography will address the question of whether the

317: bright radio sources in the fields of the two CFHQS quasars is due to foreground  

318: galaxy clusters. We notice that bright extended radio sources and overdensities were previously reported in the 1.4 GHz 

319: observations of the two SDSS z$\sim$6 quasars J1030+0524 

320: and J1148+5251 (Petric et al. 2003; Carilli et al. 2004). Carilli et al. (2004) 

321: investigated the SDSS data of the J1148+5251 field, and however, there is no evidence for 

322: dense foreground clusters or gravitational lensing. 

323: 

324: \subsection{Summary of the millimeter and radio results}

325: 

326: In summary, all the thirty-three source in the z$\sim$6 quasar 

327: sample have been observed by MAMBO at 250 GHz. The observations 

328: have rms values in the range of $\rm 0.4-1.1\,mJy$ with a median value of about $\rm

329: 0.6\,mJy$, and ten sources are detected (see Table 1 and 2).

330: This yields a detection rate of $\rm 30\pm10\%$ which is consistent 

331: with the submillimeter and millimeter detection rates of optically bright quasars at redshifts 2 and 4 (Priddey et al. 2003a; Omont

332: et al. 2001; 2003; Carilli et al. 2001).

333: 

334: Thirty-two z$\sim$6 quasars 

335: have 1.4 GHz radio continuum observations by the VLA, with a median rms level of $\rm 20\,\mu Jy$, 

336: Ten out of the thirty-two sources have been detected, including four 250 GHz detections (see Table 1 and 2). 

337: Two of the radio detections, J0836+0054 and J1427+3312, have 

338: flux densities $\rm >1mJy$. These two sources, together with the new radio detection J0203+0012,  

339: have radio loudness\footnote{The $\rm 4400\AA$ and 5 GHz flux densities are derived

340: from the 1450$\rm \AA$ magnitude and the observed 1.4 GHz flux density, 

341: assuming power law spectra indices of -0.5 in the optical 

342: and -0.75 in the radio, respectively.} 

343: $\rm R\equiv f_{\nu,5GHz}/f_{\nu,4400}\ge10$,

344: and hence are radio loud according to the

345: definition by Kellermann et al. (1989).

346: 

347: Data at other wavelengths for the analysis in this

348: paper were collected from the literature, including optical data

349: from the original papers (Fan et al. 2000; 2001a; 2003; 2004; 2006a;

350: 2008, in prep; Goto 2006; McGreer et al. 2006; Willott et al. 2007;

351: Jiang et al. 2008), near-infrared data from Jiang et al. (2006),

352: submillimeter data from Priddey et al. (2003b), Robson et al. (2004) and Beleen et al. (2006),

353: and the millimeter continuum data from the PdBI CO observations 

354: (Carilli et al. 2007; Wang et al. 2008b, in prep.).

355: 

356: \section{The UV-to-radio SEDs of the z$\sim$6 quasars}

357: 

358: Using the 250 GHz and 1.4 GHz continuum measurements, 

359: we derive the average FIR and radio emission of these 

360: z$\sim$6 quasars and compare it to the typical 

361: quasars at low redshift. The calculations

362: are performed for three groups: (I) the whole

363: sample of thirty-three sources, (II) the subsample of

364: ten sources detected with MAMBO at $\rm \ge3\sigma$, and 

365: (III) the subsample of twenty-three sources not detected with MAMBO.

366: Using the MAMBO 250GHz data, we calculate the weighted

367: average 250GHz flux densities for each group, namely:  \\ 

368: \begin{equation}

369: \rm \left\langle f_{250GHz}\right\rangle =\frac{\sum w_{i}f_{250GHz,i}}{\sum w_{i}}

370: \end{equation}

371: %\begin{equation}

372: %\rm w_{i}=\cases{\frac{1}{\sigma^{2}_{250GHz,i}} & whole\ sample;\ MAMBO\ non-detections\cr 1 & MAMBO\ detections\cr}

373: %\end{equation}

374: \begin{equation}

375: \rm w_{i}=\frac{1}{\sigma^{2}_{250GHz,i}}

376: \end{equation}

377: \begin{equation}

378: \rm \left\langle \sigma_{250GHz}\right\rangle=\left(\frac{\sum \left(w_{i}\sigma_{250GHz,i}\right)^2}{(\sum w_{i})^2}\right)^{0.5}

379: \end{equation} 

380: where $\rm f_{250GHz,i}$ is the MAMBO observed flux density for each source, 

381: and $\sigma_{250GHz,i}$ is the $\rm 1\sigma$ rms. 

382: For the calculation of $\rm \langle f_{250GHz}\rangle$ of the whole sample and the  

383: subsample of 250 GHz non-detections, we weighed the data by $\rm 1/\sigma^{2}_{250GHz,i}$ (equation (2)).

384: However, equal weights were adopted (i.e. $\rm w_{i}=1$) when calculate $\rm \langle f_{250GHz}\rangle$ 

385: for the 250 GHz detections, as the observed rms values spread in a smaller range (0.36$\sim$0.75 mJy) 

386: and we do not want to bias the average toward the weakest sources.

387: The average flux density ratios between 250GHz and 

388: the rest frame 1450$\rm \AA$ are then calculated 

389: as $\rm \langle f_{250GHz}\rangle/\langle f_{1450}\rangle$, where the average 1450$\rm \AA$ 

390: flux density $\rm \langle f_{1450}\rangle$ is derived from the AB magnitudes presented 

391: in the discovering papers (see Section 2.1).

392: %\begin{equation}

393: %\rm \left\langle \frac{f_{250GHz}}{f_{\nu,1450}}\right\rangle=\frac{\sum w_{i}\frac{f_{250GHz,i}}{f_{\nu,1450,i}}}{\sum w_{i}} \\

394: %\end{equation}

395: %\begin{equation}

396: %\rm w_{i}=(\frac{f_{\nu,1450}}{\sigma_{250GHz,i}})^{2} \\

397: %\end{equation}

398: %\begin{equation}

399: %\rm \left\langle\sigma\right\rangle=\left(\frac{1}{\sum w_{i}}\right)^{0.5} \\

400: %\end{equation}

401: The average radio emission and 1.4GHz-to-1450$\rm \AA$ ratios 

402: for each group are calculated using the same method.

403: 

404: The average FIR luminosities (42.5$\mu$m -- 122.5$\mu$m, rest frame) for each

405: group can be estimated\footnote{A typical redshift

406: value of z=6 is adopted here} as $\rm 

407: <L_{FIR}>=2.34\times10^{12}(<f_{250GHz}>/mJy)\,L_{\odot}$, assuming a

408: dust temperature of 47 K and an emissivity index $\rm \beta$ of 1.6

409: (Beelen et al. 2006). 

410: Based on this estimation, the mean

411: FIR-to-radio ratios for each group can be calculated as (Helou et al. 1985):

412: \begin{equation}

413: \rm q\equiv log\left(\frac{\langle L_{FIR}\rangle}{3.75\cdot10^{12}\,L_{\odot}}\right)-log\left(\frac{\langle L_{1.4GHz}\rangle}{L_{\odot}\cdot Hz^{-1}}\right)

414: \end{equation}

415: The rest-frame 1.4 GHz radio luminosity densities ($\rm L_{1.4GHz}$) are derived with

416: the average radio flux densities, assuming a power-law radio spectrum 

417: of $\rm f_{\nu}\sim\nu^{-0.75}$ (Condon 1992). The q values can be increased by $\sim$0.2 

418: if we assume a flater radio spectrum of $\rm f_{\nu}\sim\nu^{-0.5}$.

419: The mean flux densities,

420: luminosities, and q values are listed in Table 3. We exclude the

421: three radio loud sources in the first and the third groups, and present the corresponding

422: results in the second row of each group in Table 3. The ten MAMBO

423: detected sources in the second group are all radio quiet (according 

424: to the definition in Kellermann et al. 1989).

425: 

426: \subsection{The mean SED}

427: 

428: The mean 250 GHz flux density is $\rm 1.26\pm0.10\, mJy$ for

429: the whole sample, and $\rm 2.30\pm0.15\, mJy$ 

430: for the MAMBO detections. The mean 250 GHz 

431: flux density of MAMBO non-detections is recovered by 

432: averaging all the on-source pixel values. A 

433: $\rm \sim4\sigma$ signal of $\rm 0.52\pm0.13\, mJy$ is 

434: obtained for this group, which is about one quarter of the value of 

435: the typical MAMBO detections. We ran the stacking with 

436: the non-detection group several times, omitting one source 

437: each stack, to see if this $\rm 4\sigma$ signal is due to a

438: single source that is just below the detectability.

439: The test shows stacking flux densities in the range 

440: of 0.47 to 0.56 mJy with rms values of $\sim$0.13 mJy. Thus it 

441: is unlikely that the average flux density is dominated by one object. 

442: No significant differences are found in the $\rm \langle f_{250GHz}\rangle$ values when

443: the radio-loud sources are excluded in each group. 

444: 

445: The average 1.4 GHz observed flux densities are $\rm 27\pm4$ to $\rm 37\pm4\, \mu Jy$ 

446: for the radio quiet sources in the three groups. We convert these measurements 

447: to the flux densities at rest frame 5 GHz assuming a power-law radio spectrum 

448: of $\rm f_{\nu}\sim\nu^{-0.75}$ (Condon 1992). This gives an 

449: average radio loudness (i.e. average radio to optical emission ratio 

450: $\rm \langle R\rangle=\langle f_{\nu,5GHz}\rangle/\langle f_{\nu,4400}\rangle$) 

451: of $\rm \langle R\rangle\sim$0.7, which is consistent with the typical radio loudness values of $\rm 0.1<R<1$ 

452: found with local(Kellermann et al. 1989) and z$\leq$2 (Cirasuolo et al. 2003) radio quiet quasars.

453: The derived q parameters (i.e. FIR to radio emission ratios) are all lower or at the 

454: edge of the range of $\rm q=2.34\pm0.7$ for typical star 

455: forming galaxies (Yun et al. 2001), but are consistent with 

456: the values of q$<$2 found in the local Seyfert 1 galaxies 

457: (Ulvestad \& Ho 2001). This result suggests that, as was found in the 

458: local universe and at z$\rm \sim$2 to 4, 

459: AGN power dominates the radio emission in a large fraction of these radio quiet 

460: quasars, i.e., the radio quiet AGNs are in fact not radio silent 

461: (Barvainis et al. 2005; Carilli et al. 2001; Petric et al. 2006). 

462: 

463: We plot the mean SED (normalized by the rest frame 1450$\rm \AA$ flux density) 

464: of the radio quiet sources in each group in Figure 2, 

465: together with the templates of low-z optical quasars

466: from Elvis et al. (1994) and Richards et al. (2006).

467: We also plot the average optical to near infrared 

468: emission based on the Spitzer photometry of 13 z$\sim$6 

469: quasars from Jiang et al. (2006), who show that the ratio 

470: is consistent with the local 

471: templates. The FIR-to-1450$\rm \AA$ ratios of the three z$\sim$6 quasar groups have a 

472: range of $\sim$0.6 dex, and no obvious excess is seen between the value of the whole 

473: sample and the FIR end of the templates. 

474: However, due the lack of millimeter and submillimeter data, the available templates 

475: do not give a good constraint of the FIR-to-millimeter SED for the local 

476: optical quasars. Sanders et al. (1989) studied a sample of optically selected 

477: Palomar-Green quasars (PG quasars; Schmidt \& Green 1983; Boroson \& Green 1992), 

478: and their analysis suggest that the typical 100$\rm \,\mu$m to 1$\rm \,mm$ spectral 

479: index $\rm \alpha$ (($\rm f_{\nu}\sim\nu^{\alpha}$)) is $\rm \geq2$. We extrapolate 

480: the FIR emission of the templates with this result (an example of $\rm \alpha=2$ is plotted in Figure 2).  

481: The extrapolation matchs the mean value of the z$\sim$6 MAMBO undetected quasars very well, while 

482: the average value of the MAMBO detection is $\sim$2.5 times higher.

483: 

484: \subsection{The SEDs of objects with 250 GHz detections}

485: 

486: We plot the individual optical-to-radio spectral energy distribution

487: (SEDs) for the ten sources that were detected by MAMBO in Figure 3. 

488: The FIR emission from the two strongest MAMBO detections, J1148+5251 

489: and J0927+2001, was well measured with SHARC-II and SCUBA at shorter

490: wavelengths, which show FIR emission bumps in the SEDs with dust 

491: temperatures of 55 K and 52 K, respectively (Robson et al. 2004; 

492: Beelen et al. 2006; Wang et al. 2008a). Similar excesses are 

493: also seen in the SEDs of another five sources: J0033-0125, 

494: J0840+5624, J1335+3533, J1425+3254, and J2054-0005, where the observed 

495: 250 GHz flux densities are $\rm \gtrsim0.5$ dex higher than those 

496: expected from the quasar templates fit to the observed 

497: optical emission. 

498: 

499: We model the FIR emission of these sources with a optically thin

500: graybody assuming an emissivity index of $\rm \beta=1.6$ (Beelen et

501: al. 2006). For J0927+2001 and J1148+5251, we directly adopt the

502: dust temperatures fit in Wang et al. (2008a) and Beelen et al. (2006), respectively.

503: For the other nine sources, we assume a dust temperature of 47K (Beelen et al. 2006;

504: Wang et al. 2007). The model FIR SED was extrapolated to the radio band

505: using the typical FIR to radio correlation defined by star forming

506: galaxies, i.e. q=2.34 (Yun et al. 2001). A radio spectral index of

507: -0.75 is adopted here (Condon 1992). Three out of the seven FIR excess 

508: sources have been detected at 1.4GHz with deep VLA 

509: observations, and the observed flux densities 

510: are all above the average for their 250 GHz flux and q=2.34, 

511: but the values are within the range in which star

512: forming galaxies lie.

513: 

514: \section{Correlations between FIR emission and the AGN}

515: 

516: To investigate the origin of the FIR emission, 

517: we compare the correlation between the 

518: FIR luminosities and the central AGN bolometric luminosities of these 

519: z$\sim$6 quasars to that of two samples of local quasars: 

520: a sample of forty-five local (z$<$0.5) PG quasars  

521: from Hao et al. (2005), a sample of thirty-one z$<$0.35 type I AGNs hosted in ULIRGs

522: with IR luminosities $\rm L(8-1000\mu m)>10^{12}L_{\odot}$

523: (IR quasars; Zheng et al. 2002; Hao et al. 2005). 

524: 

525: Hao et al. (2005) reported that the IR quasar sample showed stronger FIR emission and 

526: a shallower slope in the 60 $\mu$m-optical luminosity relationship 

527: when compared to the PG quasars (see also Hao et al. 2007). 

528: They attributed this result to the starburst-dominated FIR dust heating 

529: in the host galaxies of these IR quasars. 

530: We re-fit these FIR-AGN relationships using the FIR and AGN 

531: bolometric ($\rm L_{bol}$) luminosities of the two samples. The bolometric

532: luminosities are taken from Hao et al. (2005), which were derived from the 

533: extinction-corrected quasar optical emission (see also Zheng et al. 2002). 

534: The FIR luminosities of the IR quasars are taken from Zheng et al. (2002), 

535: while for the PG quasar sample we adopt the FIR luminosities and upper limits presented

536: in Haas et al. (2003).  All these data are corrected to our cosmology.

537: %The calculation parameters adopted in 

538: %Haas et al. (2003) for the PG quasars are not exactly matched 

539: %with what we use in this paper (an emissivity index $\rm \beta=2$ is 

540: %adopted in Haas et al. 2003). However, we do not

541: %make any correction since (i) the integrated FIR luminosity is

542: %sensitive to small changes of the $\rm \beta$ value (ii) significant

543: %uncertainties of the $\rm L_{FIR}$ estimation still exist because of

544: %insufficient spectral data points for both the local and z$\sim$6

545: %quasar sample.

546: We adopt the Expectation-Maximization 

547: method in the IRAF STSDAS package (Isobe et al. 1986) which can perform linear 

548: regression with censored data. The results are: \\

549: \begin{equation}

550: \rm IR\,quasars:\,log L_{FIR}=(0.35\pm0.08)log L_{bol}+(7.9\pm1.0)

551: \end{equation}

552: \begin{equation}

553: \rm PG\,quasars:\,log L_{FIR}=(0.66\pm0.08)log L_{bol}+(2.8\pm1.0)

554: \end{equation}

555: 

556: Figure 4 compares the z$\sim$6 quasars to these two local 

557: quasar samples in the $\rm L_{FIR}-L_{bol}$ plane.

558: All the millimeter detected sources, and the average of the 

559: non-detections are presented. For millimeter detections, 

560: $\rm L_{FIR}$ was integrated 

561: using the modeled FIR SEDs described in Section 3.2 

562: (see also Wang et al. 2007), and for non-detections we 

563: adopt the estimation of average FIR luminosity in 

564: Section 3. The AGN bolometric luminosities of these z$\sim$6 

565: quasars are estimated based on optical and near-infrared 

566: observations. Jiang et al. (2006) presented $\rm L_{bol}$ for the  

567: Spitzer observed sources. We adopted their results 

568: (corrected to the cosmology we adopted in this paper) and estimated

569: $\rm L_{bol}$ from the optical B-band luminosity for the other sources,

570: $\rm L_{bol}=10.4L_{B}$ (Richards et al. 2006). The B-band 

571: luminosities are estimated with the rest frame 1450$\rm \AA$ 

572: magnitudes assuming a power-law spectrum of $\rm f_{\nu}\sim\nu^{-0.5}$. 

573: We also convert $\rm L_{bol}$ to

574: black hole accretion rate $\rm \dot M$ using\\

575: \begin{equation}

576: \rm L_{bol}= \eta \dot M c^{2},

577: \end{equation}

578: and an assumed efficiency $\eta=0.1$. The derived parameters

579: for the ten sources are listed in Table 4.

580: 

581: %The quasars at z$\sim$6 are all more luminous than are 

582: %the local quasars, and thus occupy the high luminosity region of the $\rm L_{FIR}-L_{bol}$ plot. 

583: %This is due to the detectable flux limit at z$\sim$6 

584: %and the redshift limit of the two local samples. However, it is clear that 

585: Nine of the ten MAMBO-detected z$\sim$6 quasars stand

586: close to the relationship derived from local IR quasars in the $\rm L_{FIR}-L_{bol}$ plot (Figure 4), 

587: while the other source J0818+0722 has a lower FIR luminosity, 

588: falling between the relationships of the IR and PG quasars. 

589: This may suggest that the strong millimeter detections 

590: at z$\sim$6 are the high-mass counterparts of the local 

591: IR selected quasars, and a similar starburst dust-heating 

592: mechanism dominates their FIR emission. 

593: However, these z$\sim$6 quasars are much more luminous in the optical than the 

594: two local quasars and the $\rm L_{FIR}-L_{bol}$ trends 

595: derived with the two local samples are getting mixed at 

596: high luminosity end. Thus extending the millimeter study to 

597: the optically fainter quasar population at z$\sim$6 will help to determine 

598: the $\rm L_{FIR}-L_{bol}$ correlation for the millimeter bright z$\sim$6 quasars  

599: and make a better comparison at low and high redshifts. 

600: %We present a full discussion about this in Section 6.2. 

601: On the other hand, the average value of the MAMBO non-detections is consistent 

602: with the trend defined by the local PG quasars.

603: 

604: \section{A link between millimeter detection and the strength of $\rm Ly\alpha$ emission}

605: 

606: Omont et al. (1996b) studied a sample of z$\sim$4 optically bright 

607: quasars and found that the millimeter detected sources tend to have weaker 

608: UV emission lines compared to the average line strength of the millimeter non-detections.

609: Bertoldi et al. (2003a) found a similar trend with MAMBO observations

610: of five z$\sim$6 quasars. In this section, we investigate this

611: effect with the $\rm Ly\alpha$ emission of full sample of z$\sim$6 quasars observed with MAMBO. The rest-frame $\rm Ly\alpha$

612: equivalent widths ($\rm EW_{Ly\alpha}$) of five quasars in the

613: SDSS deep strip were presented in Jiang et al. (2007a). 

614: The UV spectra of another 20 sources are available in the papers of  

615: Fan et al. (2000, 2001a, 2003, 2004, 2006a, 2008, in prep). We estimate

616: $\rm EW_{Ly\alpha}$ for 18 of them. The spectra have been 

617: corrected for Galactic extinction, adopting the extinction curve 

618: presented in Savage \& Mathis (1979). We then fit the spectra using a  

619: method similar to that used in Jiang et al. (2007a), i.e. a power-law 

620: ($\rm f_{\nu}\sim \nu^{-0.5}$) for the UV

621: continuum and gaussians for the $\rm Ly\alpha$ and $\rm NV$

622: line emission. The sources J1044-0125 and J1048+4637 

623: have broad absorption line features and their equivalent 

624: widths are therefore unavailable. 

625: 

626: We plot the histograms of $\rm EW_{Ly\alpha}$ for MAMBO detections

627: (solid line) and non-detections (dotted line) in Figure 5. 

628: Most of the MAMBO detections distribute in the region 

629: of $\rm log\,EW_{Ly\alpha}(\AA)<1.5$ with a median 

630: value of 24 $\rm \AA$, while most non-detections 

631: have $\rm log\,EW_{Ly\alpha}(\AA)$ $\rm >1.5$ with a 

632: median value of 68 $\rm \AA$ which is comparable 

633: to the average value of $\sim$70 $\rm \AA$ from the large optical 

634: quasar samples at lower redshifts (Schneider et al. 1991; Fan et al. 2001b). 

635: However, there is also one MAMBO 

636: non-detected source, J1621+5155, which shows very weak $\rm Ly\alpha$

637: emission with $\rm EW_{Ly\alpha} < 5\AA$.

638: 

639: The physical origin of this curious trend is not clear. 

640: According to the spectra study with samples of z$\sim$4 quasars presented 

641: in Omont et al. (1996b), similar weaker emission is also seen in other UV 

642: emission lines, such as the Si \textsc{\small IV}/O \textsc{\small IV]} and C \textsc{\small IV} lines. 

643: Though the UV line emission is weak in the MAMBO detected sources,

644: there is no evidence of strong dust reddening in the UV continuum.

645: We can speculate that intrinsically different physical conditions associated with the broad

646: line emission cloud, or a special dust obscuration geometry that affects 

647: only the broad line region, may explain this effect. Further tests 

648: with other strong broad lines, such as the C \textsc{\small IV}, and Mg \textsc{\small II} lines in 

649: the UV, the H$\alpha$, and H$\beta$ lines in the 

650: optical, and the Paschen lines in near-infrared, are required to

651: give a better understand of this trend. In particular, the  

652: trend may disappear when tested with the near-infrared lines if it is 

653: due to dust absorption. 

654: 

655: \section{Discussion}

656: 

657: In this paper, we analyze the FIR properties of the current quasar

658: sample at z$\sim$6. These quasars are mainly optically selected, and

659: represent the luminous end of the quasar population at this early epoch. 

660: The X-ray to near-infrared SEDs of these sources are  

661: dominated by powerful AGN, just as in  

662: optical quasars at lower redshift (Jiang et al. 2006). 

663: We divide the sample into MAMBO 250 GHz detections of 10 sources 

664: and non-detections of 23 sources. Their FIR emission was compared to 

665: that of the low-z quasars, and with the results listed above we will further discuss 

666: the FIR dust heating and evolution stage of these z$\sim$6 quasars in this section. 

667: 

668: \subsection{The average FIR emission of the 250 GHz non-detections}

669: 

670: The average FIR emission of the MAMBO non-detections is 

671: recovered at $\rm 4\sigma$ by stacking the 250 GHz 

672: measurements of each source, and the mean SED is similar 

673: to the low-z quasar templates of Elvis et al. (1994) 

674: and Richards et al. (2006). This suggests that the emission 

675: from the outer part of AGN-heated dust torus may dominates 

676: the FIR SED in these sources, as was found in the low-z 

677: optically bright quasars (Sanders et al. 1989; Haas et al. 2003).

678: However, we cannot rule out contributions from star formation in the

679: quasar hosts. Further surveys of PAH and other star forming features 

680: will address the average

681: contribution from star formation to FIR dust heating in these objects. 

682: The average FIR luminosity (42.5$\mu$m -- 122.5$\mu$m, rest frame) derived from the stacking flux 

683: density is 1.2$\rm \times10^{12}\,L_{\odot}$. 

684: If star formation is not the dominating dust heating 

685: source (i.e. contributes $<$50\%), the average star formation rate in the quasar 

686: hosts\footnote{The star formation rate is estimated with

687: the empirical relationship from Kennicutt

688: (1998), assuming a standard Salpeter initial mass 

689: function, i.e. $\rm SFR\sim 4.5L_{IR}\,M_{\odot}yr^{-1}$,

690: where $\rm L_{IR}$ is the infrared

691: luminosity (8-1000$\,\mu$m) in unit

692: of $\rm 10^{44}\,erg\,s^{-1}$, and

693: is $\rm\sim 1.5L_{FIR}$ for warm dust emission.} 

694: should be less than $\rm 200\,M_{\odot}\,yr^{-1}$.

695: %Even if we assume that 50\% of the FIR dust heating is from star formation

696: %on average, the estimated star formation rate\footnote{The star formation rate is estimated with

697: %the empirical relationship from Kennicutt

698: %(1998), assuming a standard Salpeter initial mass function, i.e. $\rm SFR\sim 4.5L_{IR}\,M_{\odot}yr^{-1}$,

699: %where $\rm L_{IR}$ is the infrared

700: %luminosity (8-1000$\,\mu$m) in unit

701: %of $\rm 10^{44}\,erg\,s^{-1}$, and

702: %is $\rm\sim 1.5L_{FIR}$ for warm dust emission.} is 

703: %only $\rm 155\,M_{\odot}\,yr^{-1}$. 

704: 

705: \subsection{The FIR emission of the strong 250 GHz detections}

706: 

707: Excess FIR emission is seen in seven of the MAMBO detected sources (see Figure 3).

708: Their 250 GHz flux densities are much stronger ($\rm \gtrsim0.5$ dex)

709: than the values expected from the quasar templates when normalized  

710: in the rest-frame UV. These sources

711: show FIR-to-radio luminosity ratios within the range defined by typical

712: star forming galaxies (Section 3.2, see also Carilli et al. 2004;

713: Wang et al. 2007). The relation between FIR and AGN bolometric luminosity for 

714: these sources follows the trend defined by a sample of low-redshift IR luminous

715: quasars which are hosted by ULIRGs (Section 4). The Spitzer

716: near-infrared photometry by Jiang et al. (2006) detected the 

717: AGN-heated hot dust ($\sim$1000 K) in some of these bright millimeter 

718: sources; the optical-to-near infrared SEDs of these sources 

719: match the low-z templates very well. The strong FIR emission bump in

720: these sources indicates an additional warm dust component, with 

721: a dust temperature of 50 to 60 K. Brightness temperature 

722: arguments imply that the warm dust emission region must be 

723: extended over a few kpc (Wang et al. 2008a). Thus either a 

724: significant fraction of the dust heating 

725: is due to star formation in the host galaxy, or the  

726: AGN must heat dust over large scales. 

727: 

728: Intense star formation activity is suggested by other observations of 

729: these strong millimeter quasars. Molecular CO emission is 

730: detected in the three strongest z$\sim$6 millimeter sources:  

731: J1148+5251 (Bertoldi et al. 2003b; Walter et al. 2003, 2004),  

732: J0927+2001 (Carilli et al. 2007), and J0840+5624 (Wang et al. 2008b, in prep).

733: These CO detections indicate the presence of 

734: $\rm \gtrsim10^{10}\,M_{\odot}$ of molecular 

735: gas in the quasar hosts (Walter et al. 2003; Bertoldi et al. 2003b; Carilli et al. 2007), 

736: which can provide the fuel required for star formation rates 

737: of up to $\rm >10^{3}\,M_{\odot}\,yr^{-1}$. 

738: Indeed, it is hard to see how the host galaxies can avoid massive 

739: star formation with so much molecular gas.

740: The measurement of several CO transitions in J1148+5251 suggests a

741: gas density of about $\rm 10^{4.5}\,cm^{-3}$ and a kinetic 

742: temperature of $\sim$100 K (Bertoldi et al. 2003b), which are

743: similar to values in starburst environments found in other quasars and

744: submillimeter galaxies at lower redshifts (Carilli et al. 2002;

745: Solomon \& Vanden Bout 2005; Riechers et al. 2006). In addition, 

746: recent high resolution observations of [C {\small II}] in J1148+5251 

747: (Walter et al. 2008, in prep.; Carilli 2008) show 

748: an emission region extended over many kpc, i.e., comparable to that of 

749: the CO (3-2) line. 

750: %The intensity of the [C II] line leads to an independent star

751: %formation rate estimate of $\rm \sim3000\,M_{\odot}\,yr^{-1}$

752: %(Maiolino et al. 2005).

753: 

754: Further studies also suggest connections between the CO and FIR 

755: emission. The FIR-to-CO luminosity ratios 

756: of J1148+5251 and J0927+2001 are about 500 and 650 

757: $\rm L_{\odot}$ $\rm (K\,km\,s^{-1}\,pc^{2})^{-1}$, respectively, which are

758: consistent with the values found in other high-z CO 

759: emitting galaxies (Solomon \& Vanden Bout 2005).

760: In Figure 6, we indicate these two CO detected z$\sim$6

761: quasars (as stars) on the FIR-CO luminosity

762: correlation plot discussed in Riechers et al. (2006).

763: The two points fall a little above but within the observed scatters of the FIR-CO luminosity relation 

764: of the typical star forming galaxies at both low and 

765: high redshifts. 

766: 

767: All these facts suggest

768: that massive star formation is an important source of

769: the FIR dust heating. However, it is difficult to constrain what fraction of the FIR

770: dust heating is contributed by star formation in these FIR

771: luminous quasars. According to the

772: FIR-to-CO luminosity correlation in Figure 6, at least $\rm 30-50\%$

773: of the FIR emission should originate from star formation, giving a 

774: corresponding star formation rate of $\rm \gtrsim 1000\,M_{\odot}\,yr^{-1}$.

775: 

776: \subsection{The evolutionary stage of the z$\sim$6 quasars}

777: 

778: Observations from optical to near-infrared of these z$\sim$6 quasars 

779: indicate that the central SMBHs have masses of $\rm \gtrsim10^{9}\,M_{\odot}$ 

780: (Willott et al. 2003; Jiang et al. 2006). 

781: Based on the  

782: SMBH-bulge mass relationship (eg. Marconi \& Hunt 2003), the final stellar 

783: bulge mass should be on order of $\rm 10^{12}\,M_{\odot}$ for a mature system. 

784: The formation of such a massive stellar system may require a 

785: star formation rate of $\rm \gtrsim10^{3}\,M_{\odot}\,yr^{-1}$

786: on time scales of a few hundred million years as was suggested 

787: by some numerical simulations (eg. Li et al. 2007a, b). 

788: %or a much lower star formation rate (i.e., $\rm \lesssim$ a few hundred $\rm M_{\odot}\,yr^{-1}$) 

789: %on longer time scales. 

790: Thus the average star formation rate 

791: of $\rm <200\, M_{\odot}\,yr^{-1}$ found in the 250 GHz 

792: non-detected quasars can be either the case that 

793: the major bulge building has already been finished 

794: and the SMBH-bulge mass relationship seen at z=0

795: has already been established, or that the host galaxy 

796: is has not yet reached the massive star formation phase,  

797: and the growth of the stellar bulge is  

798: lagging behind that of the SMBH (see eg. Shields et al. 2006; Ho 2007). 

799: Deep and high-resolution imaging at near-IR wavelengths may provide a direct measurement of 

800: the stellar population in these z$\sim$6 quasar hosts.

801: 

802: For the strong millimeter quasars at z$\sim$6, the FIR excess 

803: suggests a star formation rate of $\rm 

804: \gtrsim1000\,M_{\odot}\,yr^{-1}$ from the host galaxies. 

805: It is likely that we are witnessing simultaneous building of the 

806: stellar bulge and growth of the SMBH in these objects.

807: The case of  J1148+5251 is the best documented:

808: Willott et al. (2003) derived an estimate 

809: of $\rm 3\times10^9\,M_{\odot}$ for the black hole mass, while

810: Walter et al. (2004) estimated a dynamical mass 

811: of $\rm \sim5\times10^{10}\,M_{\odot}$. The SMBH-bulge 

812: mass ratio is thus clearly more than one order of magnitude larger than

813: the local SMBH-bulge relationship (eg. Marconi \& Hunt 2003), as in

814: several other high-z CO-detected QSOs (Shields et al. 2006; Ho 2007).

815: However, the amount of gas detected in CO observations

816: such as in J1148+5251, a few $\rm 10^{10}\,M_{\odot}$, is far short 

817: of the value required to reach the bulge mass given by the local 

818: black hole-bulge relationship.

819: 

820: To fully understand the evolutionary stage of

821: these strong millimeter quasars at z$\sim$6, further

822: observations are required. First, CO and [C

823: {\small II}] searches and excitation studies should be extended to all the 

824: z$\sim$6 quasars with strong FIR excesses to

825: determine the host galaxy ISM properties. Second, high

826: resolution ($\rm \lesssim0.3''$) CO, [C {\small II}] and dust mapping should be made

827: of all the CO detections. If the FIR-emitting

828: dust and molecular gas are distributed similarly, the FIR emission 

829: is likely to be due to a starburst, 

830: while more compact FIR emission will argue for AGN dust

831: heating. The resolved CO emission would also provide an estimate of the dynamical mass

832: of the quasar hosts on kpc scales. This is a direct, and perhaps 

833: the only, way to test the black hole-bulge relationship at 

834: the highest redshifts. These 

835: observations can be done with sensitive submillimeter,

836: millimeter and radio telescopes such as the PdBI, CARMA, and SMA, and the

837: coming ALMA and EVLA.

838: 

839: \section{Conclusion}

840: 

841: We study the SEDs of 33 quasars at z$\sim$6 from FIR to radio wavelengths. 

842: We conclude that, when averaged with the whole sample and the 

843: 250 GHz undetected sources, no significant difference is seen between the mean 

844: FIR-to-radio SEDs and the templates of low-redshift optically selected quasars.

845: In particular, we extrapolate the FIR emission of the templates with the typical 

846: quasar FIR-to-millimeter spectrum (i.e. $\rm f\sim\nu^\alpha,\,\alpha\geq2$) 

847: from Sanders et al. (1989). This extrapolation is consistent with the average FIR 

848: emission of the 250 GHz undetected z$\sim$6 quasars very well (see Section 3.1 and Figure 2). 

849: This fact suggests that these sources have a similar AGN dominated dust

850: heating mechanism. The average star formation rate from the

851: host galaxies of the quasars undetected by MAMBO is estimated to be less than a few hundred

852: $\rm M_{\odot}\,yr^{-1}$. 

853: 

854: We detect a strong FIR excess in seven of

855: the ten z$\sim$6 quasars detected in the millimeter band. These FIR luminous sources are likely

856: to be the high-mass counterparts of local IR luminous quasars which

857: are in transition between the starburst phase and the mature 

858: quasar phase. Star formation at rates of $\rm >10^{3}\,M_{\odot}\,yr^{-1}$ 

859: may be the dominant mechanism to heat the dust that gives rise to the FIR emission.

860: The millimeter detected sources also show weaker Lya emission

861: compared to that of the non-detections, but the origin of this trend is not clear yet.

862: Further observations of these FIR luminous quasars, especially spatially resolved 

863: line and dust studies, are required to further 

864: address questions of star formation and FIR dust heating in these

865: extreme objects.

866: 

867: The current sample of z$\sim$6

868: quasars is optically selected. 

869: There may be other quasar populations at z$\gtrsim$6 

870: which are still buried in their starburst environment and are obscured in

871: the optical. Sensitive IR and millimeter facilities,

872: such as Spitzer, ALMA, and Herschel, may discover such objects and provide a more 

873: complete view of galaxy and SMBH evolution in the early 

874: universe.

875: 

876: \acknowledgments 

877: We acknowledge support from the Max-Planck Society and the Alexander von Humboldt

878: Foundation through the Max-Planck-Forschungspreis 2005. 

879: Michael, A. Strauss acknowledges support of the National Science

880: Foundation grant AST-0707266. 

881: X. Fan acknowledge support from a David and Lucile Packard

882: Fellow in Science and Engineering. We thank James J. Condon for comments and suggestions. 

883: IRAM is funded by the Centre National de la Recherche Scientifique (France), 

884: the Max-Planck Gesellschaft (Germany), and the Instituto Geografico Nacional (Spain).

885: The National

886: Radio Astronomy Observatory is a facility of the National Science

887: Foundation, operated by Associated Universities, Inc.

888: 

889: Funding for the SDSS and SDSS-II has been provided by the Alfred P. 

890: Sloan Foundation, the Participating Institutions, the National Science 

891: Foundation, the U.S. Department of Energy, the National Aeronautics and 

892: Space Administration, the Japanese Monbukagakusho, the Max Planck 

893: Society, and the Higher Education Funding Council for England. The SDSS 

894: Web Site is http://www.sdss.org/.

895: The SDSS is managed by the Astrophysical Research Consortium for the 

896: Participating Institutions. The Participating Institutions are the 

897: American Museum of Natural History, Astrophysical Institute Potsdam, 

898: University of Basel, University of Cambridge, Case Western Reserve 

899: University, University of Chicago, Drexel University, Fermilab, the 

900: Institute for Advanced Study, the Japan Participation Group, Johns 

901: Hopkins University, the Joint Institute for Nuclear Astrophysics, the 

902: Kavli Institute for Particle Astrophysics and Cosmology, the Korean 

903: Scientist Group, the Chinese Academy of Sciences (LAMOST), Los Alamos 

904: National Laboratory, the Max-Planck-Institute for Astronomy (MPIA), the 

905: Max-Planck-Institute for Astrophysics (MPA), New Mexico State 

906: University, Ohio State University, University of Pittsburgh, University 

907: of Portsmouth, Princeton University, the United States Naval 

908: Observatory, and the University of Washington.

909: 

910: {\it Facilities:} \facility{IRAM:30m (MAMBO)}, \facility{VLA}, \facility{Sloan (SDSS)}

911: 

912: \begin{thebibliography}{}

913: \bibitem[]{} Abazajian, K. et al. 2003, AJ, 126, 2081

914: \bibitem[]{} Abazajian, K. et al. 2004, AJ, 128, 502

915: \bibitem[]{} Abazajian, K. et al. 2005, AJ, 129, 1755

916: \bibitem[]{} Adelman-McCarthy, J. et al. 2006, ApJS, 162, 38

917: \bibitem[]{} Adelman-McCarthy, J. et al. 2007a, ApJS, 172, 634

918: \bibitem[]{} Adelman-McCarthy, J. et al. 2008, ApJS, 175, in press

919: \bibitem[]{} Barvainis, R., Leh$\rm \acute a$r, J., Birkinshaw, M., Falcke, H., \& Blundell, K. M. 2005, ApJ, 618, 108

920: %\bibitem[]{} Beelen, A. 2004, Ph.D. thesis, Obs. Paris

921: \bibitem[]{} Beelen, A., Cox, P., Bertoldi, F. et al. 2006, ApJ, 642, 694

922: %\bibitem[]{} Bechtold, J., Siemiginowska, A. Shields, J. et al. 2003, ApJ, 588, 119

923: %\bibitem[]{} Bell, E. F. et al. 2005, ApJ, 625, 23

924: \bibitem[]{} Benford, D. J., Cox, P., Omont, A., Phillips, T. G., \& McMahon, R. G. 1999, ApJ, 518, 65

925: \bibitem[]{} Bertoldi, F., Carilli, C. L., Cox, P. et al. 2003a, A\& A, 406, L55

926: \bibitem[]{} Bertoldi, F., Cox, P., Neri, R. et al. 2003b, A\& A, 409, L47

927: \bibitem[]{} Best,P. N., Kauffmann, G., Heckman, T. M., \& Ivezi{\' c}, {\v Z}. 2005, MNRAS, 362, 9

928: \bibitem[]{} Boroson, T. A., \& Green, R. F. 1992, ApJS, 80, 109

929: %\bibitem[]{} Brandt, W. N. et al. 2002, ApJ, 569, L5

930: %\bibitem[]{} Canalizo, G., \& Stockton, A. 2001, ApJ, 555, 719

931: \bibitem[]{} Carilli, C. L., Bertoldi, F., Rupen, M. P. et al. 2001, ApJ, 555, 625

932: \bibitem[]{} Carilli, C. L. et al. 2002, AJ, 123, 1838

933: \bibitem[]{} Carilli, C. L., Walter, F., Bertoldi, F. et al. 2004, AJ, 128, 997

934: \bibitem[]{} Carilli, C. L. et al. 2007, ApJ, 666, L9

935: \bibitem[]{} Carilli, C. L. 2008, in Proceedings of Science, 'From planets to Dark ages: the modern radio Universe,' ed. Beswick

936: \bibitem[]{} Cirasuolo, M., Celotti, A., Magliocchetti, M. \& Danese, L. 2003, MNRAS, 346, 447

937: %\bibitem[]{} Cirasuolo, M., Magliocchetti, M., \& Gentile, G. et al. 2006, MNRAS, 371, 695

938: \bibitem[]{} Cool, R. J., Kochanek, C. S., Eisenstein, D. J. et al. 2006, AJ, 132, 823

939: \bibitem[]{} Condon, J. J. 1992, ARA\& A, 30, 575

940: \bibitem[]{} Condon, J. J., Cotton, W. D., Greisen, E. W., Yin, Q. F., Perley, R. A., Taylor, G. B., \& Broderick, J. J. 1998, AJ, 115, 1693

941: %\bibitem[]{} Cox, P., Beelen, A., Bertoldi, F., Omont, A., Carilli, C. L., \&

942: %Walter, F. 2005, in The Dusty and Molecular Universe, ed. A. Wilson

943: %(ESA SP-577; Noordwijk: ESA), 115

944: %\bibitem[]{} Deutsch, E. W. 1999, AJ, 118, 1882

945: \bibitem[]{} Cowie, L. L., Songaila, A., Hu, E. M., \& Cohen, J. G., 1996, AJ, 112, 839

946: %\bibitem[]{} Downes, D., Solomon, P. M. 1998, ApJ, 507, 615

947: \bibitem[]{} Dwek, E., Galliano, F., \& Jones, A. P. 2007, ApJ, 662, 927

948: \bibitem[]{} Elvis, M. et al. 1994, ApJS, 95, 1

949: \bibitem[]{} Fan, X., White, R. L., Davis, M. et al. 2000, AJ, 120, 1167

950: \bibitem[]{} Fan, X., Narayanan, V. K., Lupton, R. H. et al. 2001a, AJ, 122, 2833

951: \bibitem[]{} Fan, X. et al. 2001b, AJ, 121, 31

952: \bibitem[]{} Fan, X., Strauss, M. A., Schneider, D. P. et al. 2003, AJ, 125, 1649

953: \bibitem[]{} Fan, X., Hennawi, J. F., Richards, G. T. et al. 2004, AJ, 128, 515

954: \bibitem[]{} Fan, X., Strauss, M. A., Richards, G. T. et al. 2006a, AJ, 131, 1203

955: \bibitem[]{} Fan, X., Carilli, C. L., \& Keating, B. 2006b, ARA\&A, 44, 415

956: \bibitem[]{} Fan, X. et al. 2008, in prep.

957: \bibitem[]{} Fukugita, M., Ichikawa, T., Gunn, J. E., Doi, M., Shimasaku, K., \& Schneider, D. P. 1996, AJ, 111, 1748

958: %\bibitem[]{} Fixsen, D. H., Dwek, E., Mather, J. C., Bennett, C. L., Shafer, R. A. 1998, ApJ, 508, 123

959: %\bibitem[]{} Ferrarese, L., \& Merritt, D. 2000, ApJ, 539, L9

960: %\bibitem[]{} Fomalont, E. B. 2006, in prep.

961: %\bibitem[]{} Gebhardt, K., Bender, R., Bower, G. 2000, ApJ, 539, 13

962: %\bibitem[]{} Goto, T. 2007, AAS, 210, 0216

963: \bibitem[]{} Goto, T. 2006, MNRAS, 371, 769

964: %\bibitem[]{} Greve, T. R., Ivison, R. J., Bertoldi, F., Stevens, J. A., Dunlop, J. S., Lutz, D. \& Carilli, C. L. 2004, MNRAS, 354, 779

965: \bibitem[]{} Guilloteau, S., Omont, A., McMahon, R. G., Cox, P., \& Petitjean, P. 1997, A\&A, 328, L1

966: \bibitem[]{} Guilloteau, S., Omont, A., Cox, P., McMahon, R. G., \& Petitjean, P. 1999, A\&A 349, 363

967: %\bibitem[]{} Hass, M., M$\ddot u$ller, S. A. H., Chini, R. et al. 2000, A\& A, 354, 453

968: \bibitem[]{} Gunn, J. E. et al. 2006, AJ, 131, 2332

969: \bibitem[]{} Hao, C. N., Xia, X. Y., Mao, S. et al. 2005, ApJ, 625, 78

970: \bibitem[]{} Hao, C. N., Xia, X. Y., Mao, S., Deng, Z. G., \& Wu, H. 2007, ChJAA, submitted, (astro-ph/0704.3247)

971: \bibitem[]{} Haas, M., Klaas, U., M$\ddot u$ller, S. A. H. et al. 2003, A\& A, 402, 87

972: \bibitem[]{} Helou, G., Soifer, B. T., \& Rowan-Robinson, M. 1985, ApJ, 298, L7

973: %\bibitem[]{} Helou, G., Khan, I. R., Malek, L. et al. 1988, ApJs, 68, 151

974: \bibitem[]{} Heckman, T. M., Kauffmann, G., Brinchmann, J. et al. 2004, ApJ, 613, 109

975: %\bibitem[]{} Hildebrand, R. H. 1983, QJRAS, 24, 267

976: \bibitem[]{} Ho, L. C. 2007, ApJ, 669, 821

977: \bibitem[]{} Hogg, D. et al. 2001, AJ, 122, 2129

978: %\bibitem[]{} Hopkins, P. F., Somerville, R. S., Hernquist, L., Cox, T. J., Robertson, B., \& Li,

979: %Y. 2006, ApJ, 652, 864

980: \bibitem[]{} Hopkins, P. F., Hernquist, L., Cox, T. J., Robertson, B., \& Krause, E. 2007a, ApJ, 669, 67

981: %\bibitem[]{} Hopkins, P. F., Hernquist, L., Cox, T. J., Robertson, B., \& Krause,

982: %E. 2007, ApJ, 669, 45

983: %\bibitem[]{} Isaak, K. G., Priddey, R. S., McMahon, R. G. et al. 2002, MNRAS, 329, 149

984: \bibitem[]{} Isobe, T., Feigelson, E. D., \& Nelson, P. I. 1986, ApJ, 306, 490

985: \bibitem[]{} Ivezi{\' c}, {\v Z}. et al. 2004, Astronomische Nachrichten, 325, 583

986: %\bibitem[]{} Ivezic, Z., Richards, G., Hall, P. et al. 2004, ASPC, 311, 347

987: %\bibitem[]{} Ivezic, Z., Menou, K., Knapp, G. R. et al. 2002, AJ, 124, 2364

988: %\bibitem[]{} Iwamuro, F,. Kimura, M., Eto, S. et al. 2004, ApJ, 614, 69

989: \bibitem[]{} Jannuzi, B. T. et al. 2004, AAS, 204, 4801

990: \bibitem[]{} Jiang, L., Fan, X., Hines, D. C. et al. 2006, AJ, 132, 2127

991: \bibitem[]{} Jiang, L. et al. 2007, AJ, 134, 1150

992: \bibitem[]{} Jiang, L. et al. 2008, AJ, 2008, 135, 1057

993: %\bibitem[]{} Jiang, L., Fan, X., Ivezic, Z. et al. 2006, in prep

994: \bibitem[]{} Kauffmann, G., \& Heckman, T. M. 2005, Philos. Trans. R. Soc. London, A, 363, 621

995: \bibitem[]{} Kellermann K. I., Sramek, R., Schmidt, M. et al. 1989, AJ, 98,1195

996: \bibitem[]{} Kennicutt, R. C. 1998, ARA\&A, 36, 189

997: \bibitem[]{} Kreysa, E. et al. 1998, in Proc. SPIE Vol. 3357, p. 319-325

998: %\bibitem[]{} Magorrian, J., Tremaine, S., Richstone, D. et al. 1998, AJ, 115, 2285

999: %\bibitem[]{} Maiolino, R., Schneider, R., Oliva, E. et al. 2004, Natur., 431, 533

1000: \bibitem[]{} Li, Y. et al. 2007a, ApJ, in press, astro-ph (0608190)

1001: \bibitem[]{} Li, Y. et al. 2007b, ApJ, submitted, astro-ph (0706.3706)

1002: \bibitem[]{} Lupton, R. H., Gunn, J. E., Ivezi{\' c}, {\v Z}., Knapp, G. R., Kent, S. M., \& Yasuda, N., 2001, ASP Conf. Ser. 238: Astronomical Data Analysis Software and Systems X, 10, 269

1003: \bibitem[]{} Maiolino, R., Cox, P., Caselli, P. et al. 2005, A\& A, 440, L51

1004: \bibitem[]{} Marconi, A., \& Hunt, L. K. 2003, ApJ, 589, L21

1005: \bibitem[]{} McGreer, I. D., Becker, R. H., Helfand, D. J. et al. 2006, ApJ, 652, 157

1006: \bibitem[]{} Momjian, E., Carilli, C. L., \& Petric, A. O. 2005, AJ, 129, 1809

1007: \bibitem[]{} Momjian, E., Carilli, C. L., Riechers, D. A., Walter, F. 2007, 134, 694

1008: %\bibitem[]{} Narayanan, D. et al. 2007, ApJ, in press, astro-ph (0707.3141)

1009: %\bibitem[]{} Neistein, E., van den Bosch, F. C., \& Dekel. A. 2006, MNRAS, 372, 933

1010: %\bibitem[]{} Netzer, H. et al. 2007, ApJ, in press, astro-ph (0706.0818)

1011: \bibitem[]{} Omont, A., Petitjean, P., Guilloteau, S, McMahon, R. G., Solomon, P. M., \& P$\rm \acute{e}$contal, E. 1996a, Nature, 382, 428

1012: \bibitem[]{} Omont, A., McMahon, R. G., Cox, P., Kreysa, E., Bergeron, J., Pajot, F., \& Storrie-Lombardi, L. J. 1996b, A\&A, 315, 1

1013: \bibitem[]{} Omont, A., Cox, P., Bertoldi, F. et al. 2001, A\& A, 374, 371

1014: \bibitem[]{} Omont, A., Beelen, A., Bertoldi, F. et al. 2003, A\& A, 398, 657

1015: \bibitem[]{} Petric, A. O., Carilli, C. L., Bertoldi, F. et al. 2003, AJ, 126, 15

1016: \bibitem[]{} Petric, A. O., Carilli, C. L., Bertoldi, F. et al. 2006, AJ, 132, 1307

1017: \bibitem[]{} Pier, J. et al. 2003, AJ, 125, 1559

1018: \bibitem[]{} Priddey, R. S., Isaak, K. G., McMahon, R. G. et al. 2003a, MNRAS, 339, 1183

1019: \bibitem[]{} Priddey, R. S., Isaak, K. G., McMahon, R. G. et al. 2003b, MNRAS, 344, L74

1020: \bibitem[]{} Priddey, R. S., Ivison, R. J., \& Isaak, K. G. 2008, MNRAS, 383, 289

1021: \bibitem[]{} Richards, G. T. et al. 2006, ApJS, 166, 470

1022: \bibitem[]{} Riechers, D. A. et al. 2006, ApJ, 650, 604

1023: \bibitem[]{} Robson, I., Priddey, R. S., Isaak, K. G. et al. 2004, MNRAS, 351, L29

1024: \bibitem[]{} Sanders, D. B., Soifer, B. T., Elias, J. H., Madore, B. F., Matthews, K., Neugebauer, G., \& Scoville, N. Z. 1988, ApJ, 325, 74

1025: \bibitem[]{} Sanders, D. B., Phinney, E. S., Neugebauer, G., Soifer, B. T., \& Matthews, K. 1989, ApJ, 347, 29

1026: %\bibitem[]{} Sanders, D. B., \& Mirabel, I. F. 1996, ARA\& A, 34, 749

1027: \bibitem[]{} Savage, B. D., \& Mathis, J. S. 1979, ARA\&A, 17, 73

1028: \bibitem[]{} Schmidt, M., \& Green, R. F. 1983, ApJ, 269, 352

1029: \bibitem[]{} Schneider, D. P., Schmidt, M., \& Gunn, J. E. 1991, AJ, 101, 2004

1030: %\bibitem[]{} Schweitzer, M., Lutz, D., Sturm, E. et al. 2006, ApJ, 649, 79

1031: %\bibitem[]{} Shemmer, O., Brandt, W. N., Schneider, D. P. et al. 2006, ApJ, 644, 86

1032: \bibitem[]{} Shields, G. A., Menezes, K. L., Massart, C. A., \& Vanden Bout, P. 2006, ApJ, 641, 683

1033: \bibitem[]{} Smith, J. et al. 2002, AJ, 123, 2121

1034: \bibitem[]{} Solomon, P. M., \& Vanden Bout, P. A. 2005, ARA\&A, 43, 677

1035: \bibitem[]{} Spergel, D. N. et al. 2007, ApJS, 170, 377

1036: \bibitem[]{} Stoughton, C. et al. 2002, AJ, 123, 485

1037: \bibitem[]{} Tremaine, S., Gebhardt, K., Bender, R. et al. 2002, ApJ, 574, 740

1038: \bibitem[]{} Tucker, D. et al. 2006, Astron. Nachr., 327, 821

1039: \bibitem[]{} Ulvestad, J. S., \& Ho, L. C. 2001, ApJ, 558, 561

1040: \bibitem[]{} Venemans, B. P., McMahon, R. G., Warren, S. J., Gonzalez-Solares, E. A., Hewett, P. C., Mortlock, D. J., Dye, S., \& Sharp, R. G. 2007, MNRAS, 376, L76

1041: \bibitem[]{} Venkatesan, A., Nath, B. B., \& Shull, J. M. 2006, ApJ, 640, 31

1042: \bibitem[]{} Walter, F., Bertoldi, F., Carilli, C. L. et al. 2003, Nature, 424, 406

1043: \bibitem[]{} Walter, F., Carilli, C. L., Bertoldi, F. et al. 2004, ApJ, 615, L17

1044: \bibitem[]{} Walter, F. et al. 2008, in prep.

1045: \bibitem[]{} Wang, R., Carilli, C. L., Beelen, A. et al. 2007, AJ, 134, 617

1046: \bibitem[]{} Wang, R. et al. 2008a, AJ, 135, 1201

1047: \bibitem[]{} Wang, R. et al. 2008b, in prep.

1048: \bibitem[]{} White, R. L., Becker, R. H., Helfand, D. J. et al. 1997, ApJ, 475, 479

1049: \bibitem[]{} Willott, C. J., Steve, R., \& Grimes, J. A. 2003, ApJ, 598, 909

1050: \bibitem[]{} Willott, C. J. et al. 2007, AJ, 134, 2435

1051: \bibitem[]{} Wu, H., Zou, Z. L., Xia, X. Y., \& Deng, Z. G. 1998, A\&AS, 132, 181

1052: \bibitem[]{} York, D. G., Adelman, J., Anderson, J. E. et al. 2000, AJ, 120, 1579

1053: \bibitem[]{} Yun, M. S., Reddy, N. A., \& Condon, J. J. 2001, ApJ, 554, 803

1054: %\bibitem[]{} Zheng, Z., Wu, H., Mao, S. et al. 1999, A\& A, 349, 735

1055: \bibitem[]{} Zheng, Z., Xia, X. Y., Mao, S. et al. 2002, AJ, 124, 18

1056: \bibitem[]{} Zylka, R. 1998, MOPSI Users Manual, (IRAM: Grenoble)

1057: \end{thebibliography}

1058: 

1059: %\clearpage

1060: 

1061: \begin{table}

1062: {\scriptsize \caption{Summary of the new observations \label{tbl-1}}

1063: \begin{tabular}{lcccccc}

1064: \hline \noalign{\smallskip}

1065: \hline \noalign{\smallskip}

1066: Name & redshift &$\rm m_{1450}$ & $\rm M_{1450}$ & $\rm f_{250GHz}$ & $\rm f_{1.4GHz}$ & Reference$^{a}$ \\

1067:      &   &                 &  &   mJy            &  $\rm \mu$Jy & \\

1068: (1) & (2) & (3) & (4) & (5) & (6) & (7) \\

1069: \noalign{\smallskip} \hline \noalign{\smallskip}

1070: J000552.34$-$000655.8 & 5.85&20.23 &-26.47 & 0.36$\pm$0.48 & 40$\pm$130& 1,$\ast$,2\\

1071: J003311.40$-$012524.9 & 6.13&21.78 &-25.00 &{\bf 1.13$\pm$0.36} &-27$\pm$19 & 3,3,$\ast$\\

1072: J020332.39+001229.3 & 5.86&20.97 &-25.73 & 1.52$\pm$0.67 & {\bf 195$\pm$22} & 4,$\ast$,$\ast$\\

1073: J030331.40$-$001912.9 & 6.07&21.33 &-25.43 & 0.23$\pm$0.51 & -85$\pm$62 & 4,$\ast$,$\ast$\\

1074: J035349.76+010405.4 & 6.05&20.21 &-26.55 & 1.20$\pm$0.46 &17$\pm$19 & 4,$\ast$,$\ast$\\

1075: J081827.40+172251.8 & 6.00&19.34 &-27.40 & {\bf 1.19$\pm$0.38} &{\bf 123$\pm$12} & 5,$\ast$,2\\

1076: J084119.52+290504.5 & 5.96&19.61 &-27.12 & 1.00$\pm$0.43 &43$\pm$27 & 8,$\ast$,$\ast$\\

1077: J084229.23+121848.2 & 6.08&19.58 &-27.18 & 0.11$\pm$0.55 &-8$\pm$19 & 9,$\ast$,$\ast$\\

1078: J092721.82+200123.7 & 5.77&19.87 &-26.81 & {\bf 4.98$\pm$0.75} &{\bf 50$\pm$11} & 5,2,$\ast$\\

1079: J104433.04$-$012502.2 & 5.78&19.21 &-27.46 & {\bf 1.82$\pm$0.43} & -15$\pm$24 & 10,$\ast$,7\\

1080: J142516.30+325409.0 & 5.85&20.62 &-26.08 & {\bf 2.27$\pm$0.51} &20$\pm$20 & 14,$\ast$,$\ast$\\

1081: J142738.59+331242.0 & 6.12&20.33 &-26.44 & 0.39$\pm$0.66 &{\bf 1730$\pm$131} & 15,$\ast$,15\\

1082: J160253.98+422824.9 & 6.07&19.86 &-26.90 & 1.41$\pm$0.54 &{\bf 60$\pm$15} & 1,$\ast$,2\\

1083: J162100.70+515544.8 & 5.71&19.89 &-26.77 & 0.30$\pm$0.55 &-12$\pm$21 & 9,$\ast$,$\ast$\\

1084: J163033.90+401209.6 & 6.05&20.64 &-26.12 & 0.80$\pm$0.60 &14$\pm$15 & 11,12,$\ast$\\

1085: J164121.64+375520.5 & 6.04&21.30 &-25.45 & 0.08$\pm$0.46 &-30$\pm$32  & 3,3,$\ast$\\

1086: J205406.42$-$000514.8 & 6.07&20.67 &-26.09 & {\bf 2.38$\pm$0.53} &17$\pm$23 & 4,$\ast$,$\ast$\\

1087: J231546.36$-$002357.5 & 6.12&21.31 &-25.46 & 0.28$\pm$0.60 &31$\pm$16 & 4,$\ast$,$\ast$\\

1088: J232908.28$-$030158.8 & 6.43&21.65 &-25.20 & 0.01$\pm$0.50 &14$\pm$22 & 3,3,$\ast$\\

1089: \noalign{\smallskip} \hline

1090: \end{tabular}\\

1091: }

1092: {\scriptsize Note -- The detections are marked as boldface.\\

1093: $^{a}$The three references are for the optical, 250 GHz, 

1094: and 1.4 GHz data, respectively. The asterisks denote new data reported in this paper.\\

1095: References for Table 1 and 2 -- (1) Fan et al. 2004; (2) Wang et al. 2007; (3) Willott et al. 2007;

1096: (4) Jiang et al. 2008; (5) Fan et al. 2006a;

1097: (6) Fan et al. 2001a; (7) Petric et al. 2003; (8) Goto 2006; (9) Fan et al.

1098: 2008, in prep. (10) Fan et al. 2000; (11) Fan et al 2003; (12) Bertoldi et al.

1099: 2003a; (13) Carilli et al. 2004; (14) Cool et al. 2006; (15) McGreer et al. 2006. 

1100: }

1101: \end{table}

1102: 

1103: \begin{table}

1104: {\scriptsize \caption{Summary of previous observations \label{tbl-2}}

1105: \begin{tabular}{lcccccc}

1106: \hline \noalign{\smallskip}

1107: \hline \noalign{\smallskip}

1108: Name & redshift &$\rm m_{1450}$ & $\rm M_{1450}$ & $\rm f_{250GHz}$ & $\rm f_{1.4GHz}$ & Reference$^{a}$ \\

1109:      &   &                 &  &   mJy            &  $\rm \mu$Jy & \\

1110: (1) & (2) & (3) & (4) & (5) & (6) & (7) \\

1111: \noalign{\smallskip} \hline \noalign{\smallskip}

1112: J000239.39+255034.8 & 5.80&19.02 &-27.67 & 0.20$\pm$0.88 & {\bf 89$\pm$14} & 1,2,2 \\

1113: J083643.85+005453.3 & 5.81&18.81 &-27.88 &-0.39$\pm$0.96 &{\bf 1740$\pm$40} & 6,7,7\\

1114: J084035.09+562419.9 & 5.85&20.04 &-26.66 & {\bf 3.20$\pm$0.64} &12$\pm$9 & 5,2,2\\

1115: J103027.10+052455.0 & 6.31&19.66 &-27.16 &-1.15$\pm$1.13 & -3$\pm$20 & 6,2,2\\

1116: J104845.05+463718.3 & 6.20&19.25 &-27.55 & {\bf 3.00$\pm$0.40} &7$\pm$13 & 11,12,2\\

1117: J113717.73+354956.9 & 6.01&19.63 &-27.12 & 0.10$\pm$1.13 &9$\pm$17 & 5,2,2\\

1118: J114816.64+525150.2 & 6.42&19.03 &-27.82 & {\bf 5.00$\pm$0.60} &{\bf 55$\pm$12} & 11,12,13\\

1119: J125051.93+313021.9 & 6.13&19.64 &-27.14 & 0.07$\pm$0.90 &37$\pm$21 & 5,2,2\\

1120: J130608.26+035626.3 & 6.02&19.55 &-27.19 &-1.05$\pm$1.04 &14$\pm$21 & 6,7,7\\

1121: J133550.81+353315.8 & 5.95&19.89 &-26.84 & {\bf 2.34$\pm$0.50} &{\bf 35$\pm$10} & 5,2,2\\

1122: J141111.29+121737.4 & 5.93&19.97 &-26.75 & 1.00$\pm$0.62 &{\bf 61$\pm$16} & 1,2,2\\

1123: J143611.74+500706.9 & 5.83&20.16 &-26.54 &-0.21$\pm$1.14 & 6$\pm$16 & 5,2,2\\

1124: J150941.78$-$174926.8 & 6.12&19.82 &-26.95 & 1.00$\pm$0.46 &-- & 3,3,-\\

1125: J162331.81+311200.5 & 6.25&20.13 &-26.67 & 0.17$\pm$0.80 &24$\pm$31 & 1,2,2\\

1126: \noalign{\smallskip} \hline

1127: \end{tabular}\\

1128: }

1129: {\scriptsize Note -- The detections are marked as boldface.\\

1130: $^{a}$The three references are for the optical, 250 GHz,

1131: and 1.4 GHz data, respectively, and the corresponding literatures are listed at the end of Table 1.

1132: }

1133: \end{table}

1134: 

1135: \begin{table}

1136: {\scriptsize \caption{Mean FIR and radio emission of the z$\sim$6 quasars \label{tb3}}

1137: \begin{tabular}{lccccccc}

1138: \hline \noalign{\smallskip}

1139: \hline \noalign{\smallskip}

1140: Group & $\rm Number^{a}$ & $\rm <f_{250GHz}>$ & $\rm <L_{FIR}>$& $\rm Number^{b}$ & $\rm <f_{1.4GHz}>$&$\rm <L_{1.4GHz}>^{c}$ & q \\

1141:     &     & mJy & $\rm 10^{12}L_{\odot}$    &     & $\rm \mu$Jy &$\rm L_{\odot}\,Hz^{-1}$  & \\

1142: (1) & (2) & (3) & (4) & (5) & (6) & (7) & (8)\\

1143: \noalign{\smallskip} \hline \noalign{\smallskip}

1144: Whole sample& 33 & 1.26$\pm$0.10 & 2.9$\pm$0.2 & 32 & 46$\pm$3&0.031$\pm$0.002 &1.41$\pm$0.04 \\ 

1145: (radio quiet)& 30 & 1.29$\pm$0.10 &3.0$\pm$0.2  & 29 & 32$\pm$3 & 0.022$\pm$0.002 &1.57$\pm$0.05 \\

1146: \noalign{\smallskip} \hline \noalign{\smallskip}

1147: MAMBO detections& 10&2.73$\pm$0.06& 6.4$\pm$0.1 & 10 & 37$\pm$4 &0.024$\pm$0.003 & 1.84$\pm$0.05\\

1148: \noalign{\smallskip} \hline \noalign{\smallskip}

1149: Non-detections&23& 0.52$\pm$0.13& 1.2$\pm$0.3 & 22 & 56$\pm$4 &0.038$\pm$0.003 & 0.94$\pm$0.11\\

1150: (radio quiet) &20& 0.51$\pm$0.13& 1.2$\pm$0.3 & 19 & 27$\pm$4 &0.018$\pm$0.003 & 1.24$\pm$0.13\\

1151: %\noalign{\smallskip} \hline \noalign{\smallskip}

1152: %SDSS quasars& 27 & 1.40$\pm$0.11& 27.57$\pm$2.16 &  27 & 49$\pm$3 & 1.14$\pm$0.06 & 1.43\\

1153: %(radio quiet)& 25 & 1.42$\pm$0.11& 29.37$\pm$2.23 &  25 & 35$\pm$3 & 0.73$\pm$0.06 & 1.57\\

1154: \noalign{\smallskip} \hline

1155: \end{tabular}\\

1156: }

1157: {\scriptsize $^{a}$ Number of sources observed at 250 GHz.\\

1158: $^{b}$ Number of sources observed at 1.4 GHz.\\

1159: $^{c}$ A radio spectral index of -0.75 (Condon 1992) is adopted here to calculate the rest frame 1.4 GHz lunimosity.

1160: }

1161: \end{table}

1162: 

1163: \begin{table}

1164: {\scriptsize \caption{Luminosities of the 250 GHz detected z$\sim$6 quasars \label{tb4}}

1165: \begin{tabular}{cccc}

1166: \hline \noalign{\smallskip}

1167: \hline \noalign{\smallskip}

1168: Name &  $\rm \frac{L_{bol}}{10^{13}L_{\odot}}$&$\rm \frac{\dot M}{M_{\odot}\,yr^{-1}}$ & $\rm \frac{L_{FIR}}{10^{12}L_{\odot}}$\\

1169: (1) & (2) & (3) & (4) \\

1170: \noalign{\smallskip} \hline \noalign{\smallskip}

1171: J003311.40-012524.9   &  1.4&9.4   &   2.6$\pm$0.8\\

1172: J081827.40+172251.8   & 12.8&86.3  &   2.8$\pm$0.9\\

1173: J084035.09+562419.9   &  6.5&43.6  &   7.6$\pm$1.5\\

1174: J092721.82+200123.7   &  7.4&50.0  &  12.2$\pm$2.8\\

1175: J104433.04-012502.2   & 13.5&91.1  &   5.3$\pm$0.8\\

1176: J104845.05+463718.3   & 14.6&98.4  &   6.9$\pm$0.9\\

1177: J114816.64+525150.2   & 18.8&126.8  &  13.4$\pm$3.4\\

1178: J133550.81+353315.8   &  7.6&51.4  &   5.5$\pm$1.2\\

1179: J142516.30+325409.0   &  3.8&25.6  &   5.4$\pm$1.2\\

1180: J205406.42-000514.8   &  3.8&25.8  &   6.8$\pm$1.1\\

1181: \noalign{\smallskip} \hline

1182: \end{tabular}\\

1183: }

1184: {\scriptsize Note -- $\rm {\dot M}$ is the black hole accretion rate derived with equation (7) in Section 4.\\

1185: }

1186: \end{table}

1187: 

1188: \clearpage

1189: 

1190: \begin{figure}

1191: \plotone{f1a.eps}\\

1192: \epsscale{1.0}

1193: \caption{Radio 1.4 GHz field images ($\rm 25'.6\times25'.6$) of J0033-0125 (a) at $\rm 6''.5\times5 ''$ 

1194: resolution (FWHM) and J1641+3755 (b) at $\rm 6''\times4''.5$ resolution. 

1195: Contour levels are a 

1196: geometric progression by a factor of 2, starting at 0.120 mJy$\rm\, beam^{-1}$,

1197: and the crosses denote the optical quasar positions.}

1198: \end{figure}

1199: \begin{figure}

1200: \figurenum{1}

1201: \plotone{f1b.eps}\\

1202: \epsscale{1.0}

1203: \caption{Continued}

1204: \end{figure}

1205: 

1206: \begin{figure}

1207: \plotone{f2.eps}\\

1208: \epsscale{1.0} \caption{Mean FIR and radio emission (normalized by the rest frame $\rm 1450\AA$ luminosity density) of the

1209: radio quiet z$\sim$6 quasars. Black

1210: squares, red circles, and green triangles represent

1211: the average values of the whole z$\sim$6 sample, MAMBO 250 GHz

1212: detections, and non-detections, respectively. 

1213: We also plot the average near-infrared emission (black squares) 

1214: derived from the Spitzer photometry of 13 sources from Jiang et al. (2006). 

1215: The dashed and solid lines are the local quasar templates from Elvis et 

1216: al. (1994) and Richards et al. (2006), respectively. 

1217: The dotted line denotes an extrapolation of the template FIR emission (Elvis et al. 1994) with a power 

1218: law of $\rm L_{\nu}\sim \nu^{2}$  }

1219: \end{figure}

1220: 

1221: \begin{figure}

1222: \plottwo{f3a.eps}{f3b.eps}\\

1223: \plottwo{f3c.eps}{f3d.eps}\\

1224: \caption{Individual UV to radio SEDs of the 250 GHz detected

1225: z$\sim$6 quasars. Squares show the observational data at 1450$\rm \AA$, 

1226: 3.6 $\rm \mu$m, 4.5 $\rm \mu$m, 5.6 $\rm \mu$m, 8.0 $\rm \mu$m, 

1227: 350 $\rm \mu$m, 450 $\rm \mu$m, 850 $\rm \mu$m, 1.2 mm (250 GHz), 

1228: and 20 cm (1.4 GHz), and arrows denote 3$\rm \sigma$

1229: upper limits. The 3 mm continuum data for the sources J0927+2001 

1230: and J2054-0005 from the PdBI CO observations (Carilli et al. 2007; Wang et al. 2008b, in prep.) are also included.

1231: Local quasar templates are plotted as in Figure-2

1232: and normalized to rest frame 1450$\rm \AA$. The thick solid

1233: line is a warm dust model (see Section 3.2) normalized to the

1234: submm data and extended to the radio band

1235: with the typical radio-FIR correlation of star forming galaxies, i.e.

1236: q=2.34 (Yun et al. 2001), and the dotted lines factors of 5 

1237: excesses above and below the typical q value.}

1238: \end{figure}

1239: \begin{figure}

1240: \figurenum{3}

1241: \plottwo{f3e.eps}{f3f.eps}\\

1242: \plottwo{f3g.eps}{f3h.eps}\\

1243: %\caption{Continu}

1244: %\end{figure}

1245: %\begin{figure}

1246: %\figurenum{2}

1247: \plottwo{f3i.eps}{f3j.eps}\\

1248: \caption{Continued}

1249: \end{figure}

1250: \begin{figure}

1251: \plotone{f4.eps}\\

1252: \caption{The $L_{FIR}$--$L_{bol}$ correlation. The 250 GHz

1253: detected z$\sim$6 quasars are plotted as black open squares with error

1254: bars denoting 1$\sigma$ rms. The average value of the 250 GHz 

1255: non-detections is plotted as a gray square. 

1256: The local IR and PG quasars from Hao

1257: et al. (2005) are plotted as circles and diamonds with

1258: arrows denoting upper limits in $\rm L_{FIR}$. The dashed lines

1259: represent the linear regression results for the two local quasar

1260: samples. We convert $\rm L_{bol}$ to the mass accretion 

1261: rate ($\rm \dot M$) at the top abscissa.}

1262: \end{figure}

1263: 

1264: \begin{figure}

1265: \plotone{f5.eps}\\

1266: \caption{The distribution of rest-frame equivalent widths 

1267: of $\rm Ly\alpha$ (EW$\rm _{ly\alpha}$) for MAMBO detected (solid line)

1268: and non-detected (dotted line) quasars at z$\sim$6. The shadowed 

1269: area denotes the two sources that only have upper limits 

1270: for the $\rm Ly\alpha$ emission. The solid and dotted lines 

1271: represent the median values for MAMBO detections 

1272: and non-detections, respectively.}

1273: \end{figure}

1274: 

1275: \begin{figure}

1276: \plotone{f6.eps}\\

1277: \caption{$\rm L_{FIR}$ vs. CO luminosity ($\rm L'_{CO}$) for 

1278: galaxies and quasars at various redshifts taken from Riechers et al. (2006).

1279: The dashed line represents the relationship 

1280: $\rm L_{FIR}\propto {L'_{CO}}^{1.39}$, derived

1281: from low-z ULIRGs, starbursts, high-z submm and radio galaxies,

1282: and low-z and high-z CO detected quasars in Riechers et al. (2006).

1283: The stars represent the two CO detected quasars in the z$\sim$6 sample.}

1284: \end{figure}

1285: 

1286: 

1287: 

1288: \end{document}

1289: