0710.4525/ms.tex
1: %\documentclass[12pt,preprint]{aastex}
2: \documentclass{emulateapj}
3: 
4: \newcommand{\myemail}{riechers@mpia.de}
5: %\usepackage{amssymb}
6: 
7: %% -------Author macros----------------------
8: %% DO NOT USE IN MATH MODE! 
9: \def\ts     {\thinspace}
10: \def\kms    {\ts km\ts s$^{-1}$}
11: \def\etal   {{\rm et\ts al.}}
12: \def\msol   {$M_{\odot}$}
13: \def\lsol   {$L_{\odot}$}
14: \def\ci     {C\ts {\scriptsize I}}
15: \def\hi     {H\ts {\scriptsize I}}
16: \def\aco    {{\rm CO}($J$=1$\to$0)}
17: \def\bco    {{\rm CO}($J$=2$\to$1)}
18: \def\cco    {{\rm CO}($J$=3$\to$2)}
19: \def\dco    {{\rm CO}($J$=4$\to$3)}
20: \def\eco    {{\rm CO}($J$=5$\to$4)}
21: \def\fco    {{\rm CO}($J$=6$\to$5)}
22: \def\gco    {{\rm CO}($J$=7$\to$6)}
23: \def\hco    {{\rm CO}($J$=8$\to$7)}
24: \def\ico    {{\rm CO}($J$=9$\to$8)}
25: \def\jco    {{\rm CO}($J$=10$\to$9)}
26: \def\kco    {{\rm CO}($J$=11$\to$10)}
27: \def\cho    {{\rm H$_2$O}(3$_{13}$$\to$2$_{20}$)}
28: \def\ahcn    {{\rm HCN}($J$=1$\to$0)}
29: \def\bhcn    {{\rm HCN}($J$=2$\to$1)}
30: \def\chcn    {{\rm HCN}($J$=3$\to$2)}
31: \def\ehcn    {{\rm HCN}($J$=5$\to$4)}
32: \def\ahco    {{\rm HCO$^+$}($J$=1$\to$0)}
33: \def\bhco    {{\rm HCO$^+$}($J$=2$\to$1)}
34: \def\dhco    {{\rm HCO$^+$}($J$=4$\to$3)}
35: \def\bhnc    {{\rm HNC}($J$=2$\to$1)}
36: \def\bcch    {{\rm C$_2$H}($N$=2$\to$1)}
37: %%-------------------------------------------
38: 
39: \slugcomment{draft version \today, accepted for publication in the Astrophysical Journal Letters}
40: 
41: \shorttitle{Search for HCN(2-1) at $z$=6.42}
42: \shortauthors{Riechers et al.}
43: 
44: \begin{document}
45: 
46: \title{Observations of Dense Molecular Gas in a Quasar Host Galaxy at
47:   $z$=6.42: %\\
48:   Further Evidence for a Non-Linear Dense Gas -- Star Formation
49:   Relation at Early Cosmic Times}
50: 
51: \author{Dominik A. Riechers\altaffilmark{1}, Fabian
52:   Walter\altaffilmark{1}, Christopher L. Carilli\altaffilmark{2}, and
53:   Frank Bertoldi\altaffilmark{3} }
54: 
55: 
56: \altaffiltext{1}{Max-Planck-Institut f\"ur Astronomie, K\"onigstuhl 17, 
57: Heidelberg, D-69117, Germany; riechers@mpia.de}
58: 
59: \altaffiltext{2}{National Radio Astronomy Observatory, PO Box O,
60:   Socorro, NM 87801, USA}
61: 
62: \altaffiltext{3}{Argelander-Institut f\"ur Astronomie, Universit\"at
63:   Bonn, Auf dem H\"ugel 71, Bonn, D-53121, Germany}
64: 
65: %\email{riechers@mpia.de}
66: 
67: \begin{abstract}
68:   We report a sensitive search for the \bhcn\ emission line towards
69:   SDSS\,J114816.64+525150.3 (hereafter:\ J1148+5251) at $z$=6.42 with
70:   the Very Large Array (VLA).  HCN emission is a star formation
71:   indicator, tracing dense molecular hydrogen gas ($n({\rm H_2}) \geq
72:   10^4\,$cm$^{-3}$) within star-forming molecular clouds.  No emission
73:   was detected in the deep interferometer maps of J1148+5251.  We
74:   derive a limit for the HCN line luminosity of $L'_{\rm HCN} < 3.3
75:   \times 10^{9}\,$K \kms pc$^2$, corresponding to a HCN/CO luminosity
76:   ratio of $L'_{\rm HCN}$/$L'_{\rm CO}$$<$0.13. This limit is
77:   consistent with a fraction of dense molecular gas in J1148+5251
78:   within the range of nearby ultraluminous infrared galaxies (ULIRGs;
79:   median value:\ $L'_{\rm HCN}$/$L'_{\rm CO}$=0.17$^{+0.05}_{-0.08}$)
80:   and HCN-detected $z$$>$2 galaxies (0.17$^{+0.09}_{-0.08}$).  The
81:   relationship between $L'_{\rm HCN}$ and $L_{\rm FIR}$ is considered
82:   to be a measure for the efficiency at which stars form out of dense
83:   gas.  In the nearby universe, these quantities show a linear
84:   correlation, and thus, a practically constant average ratio.  In
85:   J1148+5251, we find $L_{\rm FIR}$/$L'_{\rm HCN}$$>$6600. This is
86:   significantly higher than the average ratios for normal nearby
87:   spiral galaxies ($L_{\rm FIR}$/$L'_{\rm HCN}$=580$^{+510}_{-270}$)
88:   and ULIRGs (740$^{+505}_{-50}$), but consistent with a rising trend
89:   as indicated by other $z$$>$2 galaxies (predominantly quasars;
90:   1525$^{+1300}_{-475}$).  It is unlikely that this rising trend can
91:   be accounted for by a contribution of active galactic nucleus (AGN)
92:   heating to $L_{\rm FIR}$ alone, and may hint at a higher median gas
93:   density and/or elevated star-formation efficiency toward the more
94:   luminous high-redshift systems.  There is marginal evidence that the
95:   $L_{\rm FIR}$/$L'_{\rm HCN}$ ratio in J1148+5251 may even exceed the
96:   rising trend set by other $z$$>$2 galaxies; however, only future
97:   facilities with very large collecting areas such as the Square
98:   Kilometre Array (SKA) will offer the sensitivity required to further
99:   investigate this question.
100: \end{abstract}
101: 
102: \keywords{galaxies: active, starburst, formation, high redshift ---
103:   cosmology: observations --- radio lines: galaxies}
104: 
105: \section{Introduction}
106: 
107: High redshift galaxy populations are now being detected back to 780
108: million years after the Big Bang (spectroscopically confirmed:\
109: $z$=6.96; Iye \etal\ \citeyear{iye06}), probing into the epoch of
110: cosmic reionization (e.g., Fan \etal\ \citeyear{fan06}; Hu \& Cowie
111: \citeyear{hu06}). Many of these very distant galaxies show evidence
112: for star formation activity (e.g., Taniguchi \etal\ \citeyear{tan05}).
113: Some are even found to be hyperluminous infrared galaxies (HLIRGs;
114: Bertoldi \etal\ \citeyear{ber03a}; Wang \etal\ \citeyear{wan07}) with
115: far-infrared (FIR) luminosities exceeding 10$^{13}$\,\lsol, suggesting
116: vigorous star formation and/or AGN activity.  To probe the earliest
117: stages of galaxy formation and the importance of AGN in this process,
118: it is necessary to study the star formation characteristics of these
119: galaxies.
120: 
121: A good diagnostic to examine the star-forming environments in distant
122: HLIRGs are observations of molecular gas, the fuel for star formation.
123: The by far brightest and most common indicator of molecular gas in
124: galaxies is line emission from the rotational transitions of carbon
125: monoxide (CO), which was detected in $\sim$40 galaxies at high
126: redshift ($z$$>$1; see Solomon \& Vanden Bout \citeyear{sv05} for a
127: review). These observations have revealed molecular gas reservoirs
128: with masses of $>$10$^{10}$\,\msol\ in these galaxies, even in the
129: highest redshift quasar known, J1148+5251 at $z$=6.42 (Walter \etal\
130: \citeyear{wal03}, \citeyear{wal04}; Bertoldi \etal\
131: \citeyear{ber03b}).
132: 
133: Although CO is a good tracer of the total amount of molecular gas in a
134: galaxy, due to the relatively low critical density of $n_{\rm H_2}
135: \sim 10^2-10^3\,$cm$^{-3}$ required to collisionally excite its lower
136: $J$ transitions, it is not a reliable tracer of the dense molecular
137: cloud cores where the actual star formation takes place.  Recent
138: studies of nearby actively star-forming galaxies have shown that
139: hydrogen cyanide (HCN) is a far better tracer of the dense ($n_{\rm
140:   H_2} \sim 10^5-10^6$\,cm$^{-3}$) molecular gas where stars actually
141: form (e.g.\ Gao \& Solomon \citeyear{gao04a}, \citeyear{gao04b},
142: hereafter:\ GS04a, GS04b).  In the local universe it was found that
143: the HCN luminosity ($L'_{\rm HCN}$) scales linearly (unlike $L'_{\rm
144:   CO}$) with the FIR luminosity ($L_{\rm FIR}$) over 7--8 orders of
145: magnitude, ranging from Galactic dense cores to ULIRGs (Wu et al.\
146: \citeyear{wu05}).  As $L_{\rm FIR}$ traces the massive star formation
147: rate (unless AGN heating is significant), this implies that HCN is
148: also a good tracer of star formation.
149: 
150: HCN has now also been detected in five galaxies at $z$$>$2 (Solomon
151: \etal\ \citeyear{sol03}; Vanden Bout \etal\ \citeyear{vdb04}; Carilli
152: \etal\ \citeyear{car05}, hereafter:\ C05; Wagg \etal\
153: \citeyear{wag05}; Gao \etal\ \citeyear{gao07}, hereafter:\ G07).
154: Adding a number of upper limits obtained for other high-$z$ galaxies,
155: these observations indicate that the more luminous, higher redshift
156: systems systematically deviate from the linear $L'_{\rm HCN}$--$L_{\rm
157:   FIR}$ correlation found in the local universe (G07), and hint at a
158: rising slope of the relation toward high $L_{\rm FIR}$ and/or $z$.  To
159: further investigate this apparent non-linear, rising trend, our aim
160: has been to extend the range of existing HCN observations beyond
161: redshift 6 and to higher $L_{\rm FIR}$.
162: 
163: In this letter, we report sensitive VLA\footnote{The Very Large Array
164:   is a facility of the National Radio Astronomy Observatory, operated
165:   by Associated Universities, Inc., under cooperative agreement with
166:   the National Science Foundation.} observations of \bhcn\ emission
167: toward the $z$=6.42 quasar J1148+5251, the highest redshift source
168: detected in CO.  A previous, less sensitive search for \bhcn\ emission
169: in this source has yielded no detection (C05).  We use a concordance,
170: flat $\Lambda$CDM cosmology throughout, with
171: $H_0$=71\,\kms\,Mpc$^{-1}$, $\Omega_{\rm M}$=0.27, and
172: $\Omega_{\Lambda}$=0.73 (Spergel \etal\ \citeyear{spe06}).
173: 
174: \section{Observations}
175: 
176: \begin{figure}
177: \epsscale{1.2}
178: \plotone{f1.eps}
179: \vspace{-5mm}
180: 
181: \caption{VLA observations of \bhcn\ ({\em left}) and 23.9\,GHz
182: continuum emission ({\em right}) towards J1148+5251 at a
183: resolution of 4.0\,$''$$\times$3.3\,$''$ ({\em left}) and
184: 3.5\,$''$$\times$3.3\,$''$ ({\em right}; as indicated in the bottom
185: left corner of both panels). The \bhcn\ line map is integrated over the
186: full 25\,MHz IF.  No emission is detected in both maps.  The cross
187: indicates the geometrical center of the CO emission.  Contours are
188: shown at (-3, -2, 2, 3)$\times\sigma$ (1$\sigma = 15\,\mu$Jy
189: beam$^{-1}$ for the line map, and 14\,$\mu$Jy beam$^{-1}$ for the
190: continuum map).
191: \label{f1}}
192: \end{figure}
193: 
194: We observed the \bhcn\ transition line ($\nu_{\rm rest} =
195: 177.2612230\,$GHz) toward J1148+5251 using the VLA in D configuration
196: in 10\,tracks on 2004 June 21 and 26 (these observations were
197: discussed by C05), and between 2007 May 01 and 13. At the target
198: redshift of 6.419, the line is shifted to 23.892873\,GHz (11.96\,mm).
199: The total integration time amounts to 80\,hr.  Observations were
200: performed in fast-switching mode using the nearby source 11534+49311
201: (at $3.4^\circ$ distance to J1148+5251) for secondary amplitude and
202: phase calibration.  Observations were carried out under very good
203: weather conditions with 26 antennas [including 9 Expanded Very Large
204: Array (EVLA) antennas in 2007]:\ the phase stability was excellent
205: (typically $<$15$^\circ$ phase rms for the longest baselines).  The
206: QSO 3C286 was observed as a primary flux calibrator.  The resulting
207: flux scale is accurate within the standard 15\% uncertainty.
208: 
209: In 2004, two 25\,MHz wide intermediate frequency bands (IFs) with
210: seven 3.125\,MHz channels each were observed simultaneously, one
211: centered at the \bhcn\ line frequency (23.8929\,GHz), and one centered
212: at 24.0430\,GHz for continuum monitoring\footnote{In fact, this is the
213:   center frequency of the \bhco\ line, which however was not detected.
214:   Neither was the continuum.}. This leads to an effective bandwidth of
215: 21.875\,MHz for both the line and continuum observations,
216: corresponding to 274\,\kms\ at 23.9\,GHz. For lines as broad as CO
217: (279\,\kms\ FWHM; Bertoldi \etal\ \citeyear{ber03b}), this setup would
218: miss only $\lesssim$15\% flux in the line wings.  In 2007,
219: observations were carried out in continuum mode, with one 25\,MHz wide
220: IF centered at the \bhcn\ line frequency (23.8929\,GHz), and one
221: 50\,MHz wide IF centered at 23.7649\,GHz to monitor the source's
222: continuum\footnote{This frequency avoids other potentially bright
223:   emission lines near the \bhcn\ frequency, such as \bhco, \bhnc, and
224:   \bcch.}.
225: 
226: For data reduction and analysis, the ${\mathcal{AIPS}}$ package was
227: used. All data were mapped using the CLEAN algorithm and `natural'
228: weighting; this results in synthesized beams of
229: 4.0\,$''$$\times$3.3\,$''$ for the line map and
230: 3.5\,$''$$\times$3.3\,$''$ ($\sim$20\,kpc at $z = 6.42$) for the
231: continuum map (see Fig.~\ref{f1}). The final rms in the integrated
232: \bhcn\ line map is 15\,$\mu$Jy beam$^{-1}$, and 14\,$\mu$Jy
233: beam$^{-1}$ in the continuum map.  Averaging both uv datasets to a
234: `high-sensitivity' continuum map leads to an rms of 11\,$\mu$Jy
235: beam$^{-1}$.
236: 
237: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
238: %%%% Tab.1: CO, HCN, and continuum obs
239: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
240: \begin{deluxetable}{ r c c c }
241: \tabletypesize{\scriptsize}
242: \tablecaption{Line luminosities in SDSS\,J1148+5251. \label{tab-1}}
243: \tablehead{
244: & $S_{\nu}$ & $L'$ & Ref. \\
245: & [$\mu$Jy] & [10$^9$\,K\,\kms\,pc$^2$] & }
246: \startdata
247: \bhcn\ & (7 $\pm$ 15)   & $<$3.3         & 1 \\
248: \aco\  & $<$360         & $<$142         & 2 \\ 
249: \cco\  & 570 $\pm$ 57   & 26.4 $\pm$ 2.6 & 3 \\ 
250: \fco\  & 2450           & 26.9 $\pm$ 2.4 & 2 \\ 
251: \gco\  & 2140           & 17.3 $\pm$ 2.4 & 2 \\ 
252: \vspace{-2mm}
253: \enddata 
254: \tablerefs{${}$[1] This work, [2] Bertoldi \etal\ (\citeyear{ber03b}), 
255: [3] Walter \etal\ (\citeyear{wal03}).
256: }
257: \tablecomments{${}$ Bracketed number indicates nondetection.}
258: \end{deluxetable}
259: 
260: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
261: 
262: \section{Results}
263: 
264: No \bhcn\ emission is detected in J1148+5251 (Fig.~\ref{f1}, left). We
265: derive a peak flux density of (7 $\pm$ 15)\,$\mu$Jy beam$^{-1}$ at the
266: source's CO position (see Table~\ref{tab-1}), setting an upper
267: limit\footnote{In this letter, we quote 2$\sigma$ limits; however,
268:   note that all conclusions hold for 3$\sigma$ limits.} of 30\,$\mu$Jy
269: to the emission line peak flux.  No continuum emission is detected at
270: and/or close to the \bhcn\ line frequency (Fig.~\ref{f1}, right). From
271: the `high-sensitivity' continuum map, we derive a peak flux density of
272: (11 $\pm$ 11)\,$\mu$Jy beam$^{-1}$ at the source's position, setting
273: an upper limit of 22\,$\mu$Jy to the 23.9\,GHz continuum flux density.
274: This is consistent with the model-predicted continuum level of
275: $\lesssim$8\,$\mu$Jy (Beelen \etal\ \citeyear{bee06}).
276: 
277: From our observations, we derive a limit to the \bhcn\ line luminosity
278: of $L'_{\rm HCN} < 3.3 \times 10^9\,$K\,\kms\,pc$^2$ (assuming a
279: HCN/CO linewidth ratio of 0.67, i.e., the average value for the four
280: $z$$>$2 HCN-detected quasars\footnote{This ratio lies between 57\% and
281:   80\% for the $z$$>$2 sample, but is up to 100\% in nearby starburst
282:   galaxies like NGC\,253 (e.g., Knudsen \etal\ \citeyear{knu07}).};
283: see Table~\ref{tab-2} for references), corresponding to 13\% of the CO
284: luminosity (Walter \etal\ \citeyear{wal03}).  Using the FIR luminosity
285: derived by Beelen \etal\ (\citeyear{bee06})\footnote{FIR luminosities
286:   estimated from modeling the sparsely sampled FIR SEDs of high-$z$
287:   galaxies are only accurate within a factor of 2.}, we find $L_{\rm
288:   FIR}$/$L'_{\rm HCN}$$>$6600.
289: 
290: 
291: \section{Analysis}
292: 
293: The nondetection of \bhcn\ in J1148+5251 at the depth of our
294: observations has several implications. For the following comparison
295: with other galaxies (for which $L'_{\rm CO}$ and $L'_{\rm HCN}$ are
296: given in the ground-state transitions), we assume that the \bhcn\ and
297: \cco\ lines are thermalized, so that $L'_{{\rm HCN}(J=2-1)}$=$L'_{{\rm
298:     HCN}(J=1-0)}$ and $L'_{{\rm CO}(J=3-2)}$=$L'_{{\rm CO}(J=1-0)}$.
299: Considering the high CO excitation in this source (Bertoldi \etal\
300: \citeyear{ber03b}) and the fact that both lines arise from low $J$
301: transitions, this is likely a valid assumption.
302: 
303: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
304: %%%% Tab.2: Ratios
305: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
306: \begin{deluxetable}{ l c c l }
307: \tabletypesize{\scriptsize}
308: \tablecaption{Luminosity ratios at high $z$ and average values. \label{tab-2}}
309: \tablehead{
310: Source & $L_{\rm FIR}$/$L'_{\rm HCN}$ & $L'_{\rm HCN}$/$L'_{\rm CO}$ & Refs. \\
311: & [\lsol/$L_l$] & & }
312: \startdata
313: SDSS\,J1148+5251      & $>$6600 & $<$0.13 & 1,2,3 \\
314: \tableline
315: Cloverleaf            & 1305$\pm$360 & 0.10$\pm$0.02 & 4,5,6 \\
316: IRAS\,F10214+4724     & 2835$\pm$855 & 0.18$\pm$0.05 & 7,8,9 \\
317: VCV\,J1409+5628       & 2615$\pm$860 & 0.09$\pm$0.03 & 9,10,3 \\
318: APM\,08279+5255\tablenotemark{a} & 1000    & 0.27    & 11,12,13 \\
319: SMM\,J16359+6612      & 1550$\pm$550 & 0.18$\pm$0.04 & 14,15,16 \\
320: \tableline
321: \vspace{-2.5mm}
322: & & & \\
323: avg.\ `normal' spiral (33) & 580$^{+510}_{-270}$ & 0.04$^{+0.01}_{-0.02}$ & 17,14\\
324: avg.\ $z$$\simeq$0 LIRG (23) & 650$^{+360}_{-290}$ & 0.07$^{+0.02}_{-0.02}$ & 17,14\\
325: avg.\ $z$$\simeq$0 ULIRG (9) & 740$^{+505}_{-50}$ & 0.17$^{+0.05}_{-0.08}$ & 17,14\\
326: avg.\ $z$$\simeq$0 `all' (33+23+9) & 680$^{+400}_{-360}$ & 0.05$^{+0.05}_{-0.02}$ & 17,14\\
327: avg.\ $z$$>$2 galaxy (5) & 1525$^{+1300}_{-475}$ & 0.17$^{+0.09}_{-0.08}$ & 4--16\\
328: \vspace{-2mm}
329: \enddata 
330: \tablenotetext{a}{
331: We adopt the extrapolated \ahcn\ luminosity from Wei\ss\ \etal\
332: (\citeyear{wei07}), derived by assuming collisional
333: excitation only (uncertainties are model-dominated). 
334: Also, we only consider the `cold, dense' gas and dust component 
335: in their model, which is assumed to give rise to the HCN emission.}
336: \tablerefs{${}$[1] This work, 
337: [2] Walter \etal\ (\citeyear{wal03}), 
338: [3] Beelen \etal\ (\citeyear{bee06}), 
339: [4] Solomon \etal\ (\citeyear{sol03}), 
340: [5] Wei\ss\ \etal\ (\citeyear{wei03}), 
341: [6] Riechers \etal\ (\citeyear{rie06a}),
342: [7] Vanden Bout \etal\ (\citeyear{vdb04}),
343: [8] D.~Downes \& P.~M.~Solomon, in prep.,
344: [9] Carilli \etal\ (\citeyear{car05}),
345: [10] Beelen \etal\ (\citeyear{bee04}),
346: [11] Wagg \etal\ (\citeyear{wag05}),
347: [12] Riechers \etal\ (\citeyear{rie06b}),
348: [13] Wei\ss\ \etal\ (\citeyear{wei07}),
349: [14] Gao \etal\ (\citeyear{gao07}),
350: [15] Kneib \etal\ (\citeyear{kne05}),
351: [16] Kneib \etal\ (\citeyear{kne04}),
352: [17] Gao \& Solomon (\citeyear{gao04b}).}
353: \tablecomments{${}$ Line luminosity unit is $L_l$=K\,\kms\,pc$^2$. Bracketed numbers are sample sizes (`all' corresponds to the full Gao \& Solomon sample). For the averaged samples, median values and 1$\sigma$ statistical errors of the cumulative distribution are given. Where not quoted in the literature, 25\% error are assumed for $L_{\rm FIR}$.}
354: \end{deluxetable}
355: 
356: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
357: 
358: 
359: \subsection{`Dense Gas Fraction':\ The $L'_{\rm HCN}$/$L'_{\rm CO}$
360:   Ratio}
361: 
362: \aco\ emission is considered a good tracer for the total amount of
363: molecular gas in a galaxy, while \ahcn\ emission is considered a good
364: tracer for the dense peaks of the molecular mass distribution. The
365: $L'_{\rm HCN}$/$L'_{\rm CO}$ ratio thus is considered a measure for
366: the dense fraction of molecular gas in a galaxy\footnote{We note that
367:   Graci\'a-Carpio \etal\ (\citeyear{gra06}) have questioned the
368:   validity of $L'_{\rm HCN}$/$L'_{\rm CO}$ as a tracer of the dense
369:   gas fraction.}.  On average, this ratio is about four to five times
370: higher in local ULIRGs compared to normal, nearby spiral galaxies
371: [median value of 0.17$^{+0.05}_{-0.08}$ (ULIRGs)
372: vs.~0.04$^{+0.01}_{-0.02}$ (spirals); see Table~\ref{tab-2}].
373: However, there does not appear to be a further increase toward the
374: even more FIR-luminous systems at $z$$>$2 (0.17$^{+0.09}_{-0.08}$ on
375: average) within the statistical uncertainties (all values rederived
376: from GS04ab; G07).  The upper limit for $L'_{\rm HCN}$/$L'_{\rm CO}$
377: in J1148+5251 is 0.13, placing it below the median but within the
378: range of values measured for ULIRGs and $z$$>$2 galaxies
379: (Fig.~\ref{f2}a).  Our observations thus confirm the finding of G07
380: that galaxies brighter than $L_{\rm FIR}$=10$^{12}$\,\lsol\ appear to
381: have a higher fraction of dense molecular gas than normal spiral
382: galaxies; however, there is no indication that this fraction rises
383: further toward the highest $L_{\rm FIR}$, or that it changes with
384: redshift.
385: 
386: 
387: \subsection{`Star Formation Law':\ The $L_{\rm FIR}$/$L'_{\rm HCN}$
388:   Ratio}
389: 
390: The FIR luminosity is thought to originate dominantly from
391: dust-reprocessed light of young massive stars, and thus to be a good
392: tracer of the star-formation rate (SFR) in a galaxy. Star formation in
393: galaxies takes place in dense molecular clouds that are traced well by
394: HCN emission, resulting in a linear correlation between the FIR and
395: HCN luminosities (Wu \etal\ \citeyear{wu05}). However, G07 have found
396: a mild increase of $L_{\rm FIR}$/$L'_{\rm HCN}$ with increasing FIR
397: luminosity between normal spiral galaxies and ULIRGs [$L_{\rm
398:   FIR}$/$L'_{\rm HCN}$=580$^{+510}_{-270}$ (spirals)
399: vs.~740$^{+505}_{-50}$ (ULIRGs) on average; see Table~\ref{tab-2}].
400: This trend appears to get stronger toward the even more FIR-luminous
401: $z$$>$2 systems ($L_{\rm FIR}$/$L'_{\rm HCN}$=1525$^{+1300}_{-475}$ on
402: average). The lower limit of $L_{\rm FIR}$/$L'_{\rm HCN}$$>$6600
403: obtained for J1148+5251 is consistent with such a rising trend toward
404: high-$z$ systems, and extends it toward higher redshift
405: (Fig.~\ref{f2}b).  Due to the fact that the high-$z$ sources also have
406: systematically higher $L_{\rm FIR}$ than the nearby galaxies, it
407: remains unclear whether we observe a rising trend with redshift or
408: with $L_{\rm FIR}$ (Fig.~\ref{f2}c), or both.
409: 
410: \begin{figure*}
411: \epsscale{1.15}
412: \plotone{f2.eps}
413: 
414: \caption{Progression of the $L'_{\rm HCN}$/$L'_{\rm CO}$ (`dense gas
415: fraction') and $L_{\rm FIR}$/$L'_{\rm HCN}$ (`star formation law')
416: ratios with redshift and $L_{\rm FIR}$. Data for nearby spirals
417: ($L_{\rm FIR}$$\lesssim$10$^{10.9}$\,\lsol\, i.e., $L_{\rm
418:   IR}$$<$10$^{11}$\,\lsol) and (U)LIRGs ($L_{\rm
419:   FIR}$$\gtrsim$10$^{10.9}$\,\lsol; GS04ab; G07), and high-$z$
420: galaxies including J1148+5251 are shown (detections labeled, see
421: Table~\ref{tab-2} for references). In the {\em left} panel,
422: down-arrows indicate upper limits for HCN, and up-arrows indicate that
423: HCN was detected, but not mapped throughout the whole galaxy. In the
424: {\em middle} and {\em right} panels, this is indicated vice versa.
425: \label{f2}}
426: \end{figure*}
427: 
428: 
429: \section{Discussion}
430: 
431: 
432: \subsection{Median Gas Density and Star Formation Efficiency}
433: 
434: Krumholz \& Thompson (\citeyear{kt07}) argue that $L_{\rm
435:   FIR}$/$L'_{\rm HCN}$ is expected to be higher for galaxies with a
436: median (molecular) gas density $n_{\rm med}$ close to or higher than
437: the critical density $n_{\rm crit}^{\rm HCN}$ required for excitation
438: of the observed HCN transition than for galaxies with lower $n_{\rm
439:   med}$.  In their case, they define star formation efficiency as the
440: fraction of the mass that is converted into stars per dynamical time
441: of the system. Note that this is different than the star formation
442: rate per unit total gas mass.  They argue that the non-linear relation
443: between $L_{\rm FIR}$ and $L'_{\rm CO}$ (e.g., Kennicutt
444: \citeyear{ken98a}, \citeyear{ken98b}; GS04b; Riechers \etal\
445: \citeyear{rie06b}) arises due to the fact that CO traces all gas.  The
446: star-formation rate is then dictated by the density $n$ divided by the
447: free-fall time $\tau_{\rm ff}$ ($\tau_{\rm ff} \propto n^{-0.5}$),
448: giving the standard Schmidt-law: star formation rate $\propto
449: n^{1.5}$, or $L_{\rm FIR}$ $\propto (L'_{\rm CO})$$^{1.5}$.  For
450: molecules like HCN, which only trace the small fraction of dense gas
451: clouds directly associated with star formation in normal galaxies,
452: $\tau_{\rm ff}$ is roughly fixed by $n_{\rm crit}$.  Hence the star
453: formation rate shows a linear relationship with $n$, or $L_{\rm FIR}$
454: $\propto (L'_{\rm HCN})$$^{1.0}$. However, in extreme galaxies, where
455: $n_{\rm med}$ in the molecular ISM approaches $n_{\rm crit}^{\rm
456:   HCN}$, $\tau_{\rm ff}$ again becomes relevant (i.e., HCN emission no
457: longer selects just the rare, dense peaks whose density is fixed by
458: $n_{\rm crit}^{\rm HCN}$, but instead traces the bulk of the ISM,
459: whose density can vary from galaxy to galaxy, and thus the variation
460: of $n$ and $\tau_{\rm ff}$ re-enter the calculation), and the
461: relationship approaches $L_{\rm FIR}$ $\propto (L'_{\rm HCN})$$^{1.5}$
462: (and $L'_{\rm HCN}$ $\propto L'_{\rm CO}$).  Interestingly, current
463: data show a marginal trend for a changing power-law index at the
464: highest luminosities of the type proposed by Krumholz \& Thompson.
465: This change in power-law index from 1 to 1.5 would suggest that, in
466: these extreme luminosity systems, $n_{\rm med}$ approaches $n_{\rm
467:   crit}^{\rm HCN}$.  More systems at high luminosity are required to
468: confirm this trend of changing power-law index.
469: 
470: 
471: \subsection{The Role of AGN Heating for $L_{\rm FIR}$}
472: 
473: Like most of the $z$$>$2 HCN-detected sources, J1148+5251 is a quasar.
474: It has been found that, even for such strong AGN galaxies, the bulk of
475: $L_{\rm FIR}$ is likely dominatly heated by star formation in most
476: cases (e.g., Carilli \etal\ \citeyear{car01}; Omont \etal\
477: \citeyear{omo01}; Beelen \etal\ \citeyear{bee06}; Riechers \etal\
478: \citeyear{rie06b}). However, based on radiative transfer models of the
479: dust SED of J1148+5251, Li \etal\ (\citeyear{li07}) argue that this
480: source may currently undergo a `quasar phase', in which AGN heating of
481: the hot and warm dust contributes significantly to $L_{\rm FIR}$. If
482: correct, this may be an alternative explanation for the elevated
483: $L_{\rm FIR}$/$L'_{\rm HCN}$ in this galaxy. The (rest-frame) IR
484: properties (tracing emission from hot dust) of J1148+5251 are similar
485: to those of other $z$$>$6 quasars with much lower $L_{\rm FIR}$
486: (tracing emission from warm dust), and even to local quasars (Jiang
487: \etal\ \citeyear{jia06}).  This supports the assumption that the hot
488: dust in J1148+5251 is dominantly heated by the AGN; however, the lack
489: of a correlation between $L_{\rm IR}$ and $L_{\rm FIR}$ in quasars
490: indicates that the warm dust may still be dominantly heated by star
491: formation.  Moreover, J1148+5251 follows the radio-FIR correlation for
492: star-forming galaxies (Carilli \etal\ \citeyear{car04}), which also
493: suggests a starburst origin for the dominant fraction of $L_{\rm
494:   FIR}$.
495: 
496: Furthermore, one of the $z$$>$2 HCN detections and some of the
497: meaningful limits are submillimeter galaxies without a known luminous
498: AGN, but are still offset from the local $L_{\rm FIR}$/$L'_{\rm HCN}$
499: relation.  It thus appears unlikely that AGN heating alone can account
500: for the higher average $L_{\rm FIR}$/$L'_{\rm HCN}$ in the high-$z$
501: galaxy sample.
502: 
503: \subsection{Implications for Future Studies}
504: 
505: Even when assuming the highest $L_{\rm FIR}$/$L'_{\rm HCN}$ of 2835
506: found among all HCN-detected galaxies in Table \ref{tab-2}, the depth
507: of our observations is sufficient to detect a galaxy with the redshift
508: and $L_{\rm FIR}$ of J1148+5251 ($z$=6.42) in HCN emission at a
509: signal-to-noise ratio of $>$4.5. To first order, our lower limit thus
510: is consistent with previous suggestions (G07) that $L_{\rm
511:   FIR}$/$L'_{\rm HCN}$ ratios in high redshift sources lie
512: systematically above those for nearby galaxies.  The scatter around
513: this trend is still significant, and will primarily be improved by
514: increasing the number of HCN-detected galaxies at high $z$.  In
515: addition, it will be important to improve on the main sources of error
516: for the individual high-$z$ detections (e.g., signal-to-noise limited
517: HCN/CO linewidth ratio, accuracy of the FIR SED fit, AGN bias of
518: $L_{\rm FIR}$).  The statistical and individual results, so far, would
519: even be consistent with an even stronger increase in $L_{\rm
520:   FIR}$/$L'_{\rm HCN}$ toward the highest $z$ and/or $L_{\rm FIR}$.
521: Our study of J1148+5251 may hint at such an effect.  Clearly, it is
522: desirable to obtain more sensitive observations of this source to
523: further investigate this issue.  Due to its superior collecting area
524: and high calibrational stability, the VLA is ideally suited for such a
525: sensitive study.  Although J1148+5251 is the most CO- and FIR-luminous
526: $z$$>$6 galaxy known, 80\,hr of VLA observations were necessary to
527: obtain the current limit.  In a favourable case, the \bhcn\ line may
528: have a strength of about 1.5 times the current rms. To obtain a solid
529: 5\,$\sigma$ detection of such a line, of order 1000\,hr of
530: observations with the VLA would be required.  Due to improved
531: receivers and antenna performance, the fully operational EVLA will be
532: by a factor of two more sensitive to spectral lines of several
533: 100\,\kms\ width (such as in J1148+5251), but will still require long
534: integration times. Studies of dense gas at $z$$>$6 thus appear to
535: require an order of magnitude increase in collecting area, such as
536: offered by future facilities like the SKA phase I demonstrator (e.g.,
537: Carilli \citeyear{car06}), which can serve as a low frequency
538: counterpart to the Atacama Large Millimeter/submillimeter Array
539: (ALMA).
540: 
541: \acknowledgments 
542: The authors would like to thank Philip Solomon and Yu Gao for access
543: to their dataset of nearby galaxies. We also thank Axel Wei\ss\ for
544: helpful discussions.  D.~R. thanks Vernesa Smol{\v c}i\'c for a great
545: piece of code.  D.~R.\ acknowledges support from the Deutsche
546: Forschungsgemeinschaft (DFG) Priority Program 1177.  C.~C.\
547: acknowledges support from the Max-Planck-Gesellschaft and the
548: Alexander von Humboldt-Stiftung through the Max-Planck-Forschungspreis
549: 2005. We thank the referee for helpful comments.
550: 
551: \begin{thebibliography}{}
552: 
553: \bibitem[Beelen et al.(2004)]{bee04} Beelen, A., Cox, P., Pety, J.,
554:   \etal\ 2004, A\&A, 423, 441
555: \bibitem[Beelen et al.(2006)]{bee06} Beelen, A., Cox, P., Benford,
556:   D.~J., \etal\ 2006, ApJ, 642, 694
557: \bibitem[Bertoldi et al.(2003a)]{ber03a} Bertoldi, F., Carilli, C.~L.,
558:   Cox, P., \etal\ A\&A, 406, L55
559: \bibitem[Bertoldi et al.(2003b)]{ber03b} Bertoldi, F., Cox, P., Neri,
560:   R., et al.\ 2003b, A\&A, 409, L47
561: \bibitem[Carilli et al.(2001)]{car01} Carilli, C.~L., Bertoldi, F.,
562:   Rupen, M.~P., \etal\ 2001, ApJ, 555, 625
563: \bibitem[Carilli et al.(2004)]{car04} Carilli, C.~L., Walter, F.,
564:   Bertoldi, F., \etal\ 2004, AJ, 128, 997
565: \bibitem[Carilli et al.(2005)]{car05} Carilli, C.~L., Solomon, P.~M.,
566:   Vanden Bout, P.~A., \etal\ 2005, ApJ, 618, 586 [C05]
567: \bibitem[Carilli(2006)]{car06} Carilli, C.~L.\ 2006, SKA memo 70
568: \bibitem[Fan et al.(2006)]{fan06} Fan, X., Carilli, C.~L., \& Keating,
569:   B.\ 2006, ARA\&A, 44, 415
570: \bibitem[Gao \& Solomon(2004a)]{gao04a} Gao, Y., \& Solomon, P.~M.\
571:   2004a, ApJS, 152, 63 [GS04a]
572: \bibitem[Gao \& Solomon(2004b)]{gao04b} Gao, Y., \& Solomon, P.~M.\
573:   2004b, ApJ, 606, 271 [GS04b]
574: \bibitem[Gao et al.(2007)]{gao07} Gao, Y., Carilli, C.~L., Solomon,
575:   P.~M., \& Vanden Bout, P.~A.\ 2007, ApJ, 660, L93 [G07]
576: \bibitem[Graci\'a-Carpio et al.(2006)]{gra06} Graci\'a-Carpio, J.,
577:   Garc\'ia-Burillo, S., Planesas, P., \& Colina, L.\ 2006, ApJ, 640,
578:   L135
579: \bibitem[Hu \& Cowie(2006)]{hu06} Hu, E.~M., \& Cowie, L.~L.\ 2006,
580:   Nature, 440, 1145
581: \bibitem[Iye et al.(2006)]{iye06} Iye, M., \etal\ 2006, Nature, 443,
582:   186
583: \bibitem[Jiang et al.(2006)]{jia06} Jiang, L., Fan, X., Hines, D.~C.,
584:   \etal\ 2006, AJ, 132, 2127
585: \bibitem[Kennicutt(1998a)]{ken98a} Kennicutt, R.~C. 1998a, ApJ, 498,
586:   541
587: \bibitem[Kennicutt(1998b)]{ken98b} Kennicutt, R.~C. 1998b, ARA\&A, 36,
588:   189
589: \bibitem[Kneib et al.(2004)]{kne04} Kneib, J.-P., van der Werf, P.~P.,
590:   Knudsen, K.~K., \etal\ 2004, MNRAS, 349, 1211
591: \bibitem[Kneib et al.(2005)]{kne05} Kneib, J.-P., Neri, R., Smail, I.,
592:   \etal\ 2005, A\&A, 434, 819
593: \bibitem[Knudsen et al.(2007)]{knu07} Knudsen, K.~K., Walter, F.,
594:   Weiss, A., Bolatto, A., Riechers, D.~A., \& Menten, K.\ 2007, ApJ,
595:   666, 156
596: \bibitem[Krumholz \& Thompson(2007)]{kt07} Krumholz, M.~R., \&
597:   Thompson, T.~A.\ 2007, ApJ, in press (arXiv:0704.0792)
598: \bibitem[Li et al.(2007)]{li07} Li, Y., Hopkins, P.~F., Hernquist, L.,
599:   \etal\ 2007, ApJ, submitted (arXiv:0706.3706)
600: \bibitem[Omont et al.(2001)]{omo01} Omont, A., Cox, P., Bertoldi, F.,
601:   \etal\ 2001, A\&A, 374, 371
602: \bibitem[Riechers et al.(2006a)]{rie06a} Riechers, D.~A., Walter, F.,
603:   Carilli, C.~L., \etal\ 2006a, ApJ, 645, L13
604: \bibitem[Riechers et al.(2006b)]{rie06b} Riechers, D.~A., Walter, F.,
605:   Carilli, C.~L., \etal\ 2006b, ApJ, 650, 604
606: \bibitem[Solomon et al.(2003)]{sol03} Solomon, P., Vanden Bout, P.,
607:   Carilli, C., \& Guelin, M.\ 2003, Nature 426, 636
608: \bibitem[Solomon \& Vanden Bout(2005)]{sv05} Solomon, P.~M., \& Vanden
609:   Bout, P.~A.\ 2005, ARA\&A, 43, 677
610: \bibitem[Spergel et al.(2007)]{spe06} Spergel, D.~N., Bean, R.,
611:   Dor\'e, O., \etal\ 2007, ApJS, 170, 377
612: \bibitem[Taniguchi et al.(2005)]{tan05} Taniguchi, Y., Ajiki, M.,
613:   Nagao, T., \etal\ 2005, PASJ, 57, 165
614: \bibitem[Vanden Bout et al.(2004)]{vdb04} Vanden Bout, P., Solomon,
615:   P., \& Maddalena, R.\ 2004, ApJ, 614, L97
616: \bibitem[Wagg et al.(2005)]{wag05} Wagg, J., Wilner, D.~J., Neri, R.,
617:   Downes, D., \& Wiklind, T.\ 2005, ApJ, 634, L13
618: \bibitem[Walter et al.(2003)]{wal03} Walter, F., Bertoldi, F.,
619:   Carilli, C.~L., et al.\ 2003, Nature, 424, 406
620: \bibitem[Walter et al.(2004)]{wal04} Walter, F., Carilli, C.,
621:   Bertoldi, F., et al.\ 2004, ApJ, 615, L17
622: \bibitem[Wang et al.(2007)]{wan07} Wang, R., Carilli, C.~L., Beelen,
623:   A., \etal\ 2007, AJ, 134, 617
624: \bibitem[Weiss et al.(2003)]{wei03} Wei\ss, A., Henkel, C., Downes,
625:   D., \& Walter, F.\ 2003, A\&A, 409, L41
626: \bibitem[Weiss et al.(2007)]{wei07} Wei\ss, A., Downes, D., Neri, R.,
627:   \etal\ 2007, A\&A, 467, 955
628: \bibitem[Wu et al.(2005)]{wu05} Wu, J., Evans II, N.~J., Gao, Y.,
629:   Solomon, P.~M., Shirley, Y.~L., \& Vanden Bout, P.~A.\ 2005, ApJ,
630:   635, L173
631: 
632: \end{thebibliography}
633: 
634: \end{document}
635: