1: \documentclass[12pt,preprint]{aastex}
2: %\documentclass{aastex}
3:
4: %\usepackage{emulateapj5}
5: %\usepackage{onecolfloat}
6: %\usepackage{apjfonts}
7: %\usepackage{epsfig}
8:
9: \def\bl{\par\vskip 12pt\noindent}
10: \def\bll{\par\vskip 24pt\noindent}
11: \def\blll{\par\vskip 36pt\noindent}
12: \def\rot{\mathop{\rm rot}\nolimits}
13: \def\alf{$\alpha$}
14: \def\refff{\leftskip20pt\parindent-20pt\parskip4pt}
15: \def\zabs{$z_{\rm abs}$}
16: \def\zem{$z_{\rm em}$~}
17: \def\lya{Lyman-$\alpha$ }
18: \def\lyb{Lyman-$\beta$ }
19: \def\h2{H$_2$~}
20: \def\hi{H~{\sc i}~}
21: \def\ovi{O~{\sc vi}~}
22: \def\ovii{O~{\sc vii}~}
23: \def\oviii{O~{\sc viii}~}
24: \def\nv{N~{\sc v}~}
25: \def\niii{N~{\sc iii}$\lambda$989~}
26: \def\ovia{O~{\sc vi}$\lambda$1031~}
27: \def\ovib{O~{\sc vi}$\lambda$1037~}
28: \def\oviab{O~{\sc vi}$\lambda\lambda$1031,1037~}
29: \def\ciii{C~{\sc iii}$\lambda$977~}
30: \def\cii{C~{\sc ii}~}
31: \def\ciis{C~{\sc ii*}~}
32: \def\mgii{Mg~{\sc ii}~}
33: \def\civ{C~{\sc iv}~}
34: \def\civa{C~{\sc iv}$\lambda$1548~}
35: \def\civb{C~{\sc iv}$\lambda$1550~}
36: \def\nva{N~{\sc v}$\lambda$1238~}
37: \def\nvb{N~{\sc v}$\lambda$1242~}
38: \def\civab{C~{\sc iv}$\lambda\lambda$1548,1550~}
39: \def\sii{S~{\sc ii}~}
40: \def\alii{Al~{\sc ii}~}
41: \def\aliii{Al~{\sc iii}~}
42: \def\aliiia{Al~{\sc iii}$\lambda$1854~}
43: \def\aliiib{Al~{\sc iii}$\lambda$1862~}
44: \def\siiis{Si~{\sc ii*}~}
45: \def\siii{Si~{\sc ii}~}
46: \def\siiii{Si~{\sc iii}$\lambda$1206~}
47: \def\siiv{Si~{\sc iv}~}
48: \def\siiva{Si~{\sc iv}$\lambda$1393~}
49: \def\siivb{Si~{\sc iv}$\lambda$1402~}
50: \def\siivab{Si~{\sc iv}$\lambda\lambda$1393,1402~}
51: \def\feb{Fe~{\sc ii}$\lambda$2382~}
52: \def\febb{Fe~{\sc ii}$\lambda\lambda$2596,2600~}
53: \def\mgiia{Mg~{\sc ii}$\lambda$2796~}
54: \def\mgiib{Mg~{\sc ii}$\lambda$2803~}
55: \def\mgiiab{Mg~{\sc ii}$\lambda\lambda$2796,2803~}
56: \def\wr{$w_{\rm r}$}
57: \def\wobs{$w_{\rm obs}$}
58: \def\feb{Fe~{\sc ii}$\lambda\lambda$2586,2600~}
59: \def\kms{km~s$^{-1}$}
60: \def\tspin{T$_{\rm s}$}
61: \def\nh{n$_{\rm H}$}
62:
63: \newenvironment{inlinefigure}{%
64: \def\@captype{figure}%
65: \noindent\begin{minipage}{0.999\linewidth}\begin{center}}
66: {\end{center}\end{minipage}\smallskip}
67: \makeatother
68:
69: \begin{document}
70: %\twocolumn[%
71:
72: %\submitted{Submitted to ApJL}
73: \slugcomment{Accepted by ApJL}
74: \title{Molecular Hydrogen in a Damped Lyman-$\alpha$ System
75: at $z_{\rm abs}=4.224$}
76:
77: \author{C. Ledoux\altaffilmark{1}, P. Petitjean\altaffilmark{2,3},
78: R. Srianand\altaffilmark{4}
79: }
80:
81: \begin{abstract}
82:
83: We present the direct detection of molecular hydrogen at the highest redshift
84: known today ($z_{\rm abs}=4.224$) in a Damped Lyman-$\alpha$ (DLA) system
85: toward the quasar PSS J\,1443$+$2724. This absorber is remarkable for having
86: one of the highest metallicities amongst DLA systems at $z_{\rm abs}>3$, with a
87: measured iron abundance relative to Solar of $-1.12\pm 0.10$. We provide for
88: the first time in this system accurate measurements
89: of N\,{\sc i}, Mg\,{\sc ii}, S\,{\sc ii} and Ar\,{\sc i} column densities.
90: The sulfur and nitrogen abundances relative to Solar, $-0.63\pm 0.10$
91: and $-1.38\pm 0.10$ respectively, correspond exactly to the primary nitrogen
92: production plateau. H$_2$ absorption lines are detected in four
93: different rotational levels (${\rm J}=0$, 1, 2 and 3) of the vibrational
94: ground-state in three velocity components with total column densities
95: of $\log N({\rm H}_2)=17.67$, 17.97, 17.48 and 17.26 respectively. The
96: ${\rm J}=4$ level is tentatively detected in the strongest component
97: with $\log N({\rm H}_2)\sim 14$. The mean molecular fraction
98: is $\log f=-2.38\pm 0.13$, with
99: $f=2N({\rm H}_2)/(2N({\rm H}_2)+N($H\,{\sc i}$))$. We also
100: measure $\log N($HD$)/N($H$_2)<-4.2$. The excitation temperatures $T_{01}$ for
101: the two main components of the system are 96 and 136~K respectively. We
102: argue that the absorbing galaxy, whose star-formation activity must
103: have started at least $2-5\times 10^8$~yrs before $z=4.224$, is in a
104: quiescent state at the time of observation. The density of the gas is small,
105: $n_{\rm H}\le 50$~cm$^{-3}$, and the temperature is of the order
106: of $T\sim 90-180$~K. The high excitation of neutral carbon in one of
107: the components can be explained if the temperature of the Cosmic Microwave
108: Background Radiation has the value expected at the absorber redshift,
109: $T=14.2$~K. These observations demonstrate the feasibility to study H$_2$
110: at the highest redshifts provided high enough spectral resolution and
111: good S/N ratio are achieved.
112:
113: \end{abstract}
114:
115: \keywords{cosmology: observations -- galaxies: formation -- ISM: molecules
116: -- quasars: absorption lines}
117:
118: %]
119:
120: \altaffiltext{1}{European Southern Observatory,
121: Alonso de C\'ordova 3107, Casilla 19001, Vitacura, Santiago, Chile;
122: E-mail: cledoux@eso.org}
123: \altaffiltext{2}{Institut d'Astrophysique de Paris -- CNRS,
124: 98bis Boulevard Arago, F-75014 Paris, France;
125: E-mail: petitjean@iap.fr}
126: \altaffiltext{3}{LERMA, Observatoire de Paris,
127: 61 Avenue de l'Observatoire, F-75014, Paris, France}
128: \altaffiltext{4}{IUCAA, Post Bag 4, Ganesh Khind, Pune 411\,007, India;
129: E-mail: anand@iucaa.ernet.in}
130:
131: \section{Introduction}
132:
133: Although molecular hydrogen, H$_2$, is the most abundant mo\-le\-cu\-le in the
134: Universe, it is very difficult to detect directly. Emission lines from the
135: tracer molecule CO are usually used instead but, at $z>4$, only the
136: brightest objects can be detected with current facilities
137: (e.g., Omont et al. 1996; Solomon \& Vanden Bout 2005). For the time being,
138: the only way to detect H$_2$ directly at high redshift is to search for the
139: absorption signature of the molecular UV transition lines from
140: Damped Lyman-$\alpha$ (DLA) systems observed in the spectra of quasars or
141: those of $\gamma$-ray burst afterglows.
142:
143: DLA systems are characterized by very large neutral hydrogen column
144: densities, $N($H\,{\sc i}$)\ga 10^{20}$ cm$^{-2}$, that are typical of
145: sightlines through local spiral galaxies. Inferred metallicities typically
146: vary between [Zn,S/H$]=-2$ and -0.5 at $2\la z_{\rm abs}\la 3$ and are smaller
147: at larger redshifts (e.g., Prochaska et al. 2003). The dust content is less
148: than or about 10\% of what is observed in the Galactic ISM for
149: similar H\,{\sc i} column densities (Ledoux et al. 2003). The detection of
150: H$_2$ molecules in high-redshift DLA systems through Lyman-Werner band
151: absorption is observationally challenging due to the presence of the Ly$\alpha$
152: forest. High spectral resolution, good S/N ratio and large wavelength
153: coverage are simultaneously needed to detect the H$_2$ lines and avoid
154: spurious detections (e.g.,
155: Levshakov \& Varshalovich 1985; Ge \& Bechtold 1999; Petitjean et al. 2000; Levshakov et al. 2002; Ledoux et al. 2002, 2003; Cui et al. 2005).
156:
157: We are conducting a survey for molecular hydrogen in DLA systems at
158: high redshift ($z_{\rm abs}>1.8$; see Ledoux et al. 2003), using the
159: Ultraviolet and Visible Echelle Spectrograph (UVES; Dekker et al. 2000) at
160: the European Southern Observatory's Very Large Telescope, down to a
161: detection limit of typically $N({\rm H}_2)=2\times 10^{14}$ cm$^{-2}$. Out
162: of the 75 systems observed up to now, 14 have firm detections of associated
163: H$_2$ absorption lines. In this Letter, we present the highest redshift
164: detection in our sample, in the $z_{\rm abs}=4.224$ DLA system toward
165: the quasar PSS J\,1443$+$2724. We describe the observations in Sect.~2 and the
166: analysis of the data in Sect.~3. We conclude in Sect.~4.
167:
168: \section{Observations}
169:
170: The quasar PSS J\,1443$+$2724 ($z_{\rm em}=4.42$, $r\sim 19.3$) was observed
171: with UVES in service mode on March 2-3 and 17-19, 2004. The red
172: spectroscopic arm of UVES was used in the standard configuration with
173: the central wavelength adjusted to 580~nm. The wavelength coverage
174: was 478$-$681~nm with only a small gap, between 575 and 584~nm,
175: corresponding to the physical gap between the two red arm CCDs. The CCD pixels
176: were binned $2\times 2$ and the slit width was fixed to $1\arcsec$,
177: yielding, under the typical $0\farcs 7$ seeing conditions achieved during the
178: observations, a resolving power $R\approx 48000$. The total on-source
179: integration time was 7.3~h, split into five exposures. The data were
180: reduced using the UVES pipeline (Ballester et al. 2000) which is available as
181: a dedicated context of the ESO MIDAS data reduction system.
182: %The pipeline products were checked step by step.
183: The wavelength scale of the spectra
184: reduced by the pipeline was converted to vacuum-heliocentric values
185: and individual exposures were scaled, weighted and combined altogether to
186: produce the final object flux and variance spectra.
187:
188: \section{Analysis}
189:
190: \subsection{The high metallicity and the star-formation history}
191:
192: The metallicity of the DLA system toward PSS J\,1443$+$2724 is the
193: highest among the ten $z>4$ DLA systems investigated up to now
194: (Prochaska et al. 2003). Total integrated column densities and metallicities
195: for N\,{\sc i}, Mg\,{\sc ii}, S\,{\sc ii}, Ar\,{\sc i} and Fe\,{\sc ii},
196: obtained from Voigt-profile fitting of the observed transition lines, are given
197: in Table~\ref{Tabmetal}. With the exception of Fe\,{\sc ii}, these ions are
198: observed here for the first time. We derive metallicities relative to Solar
199: of [S/H$]=-0.63\pm 0.10$ and [Fe/H$]=-1.12\pm 0.10$. The latter value is
200: consistent with the previous measurement by Prochaska et al. (2001) but we
201: note that the newly determined H\,{\sc i} and Fe\,{\sc ii} column densities
202: are both larger by about 0.15~dex. Even though the depletion factor is moderate
203: (i.e., [S/Fe$]=+0.49$), because of the high metallicity the dust-to-gas
204: ratio measured in this system ($\log\kappa =-0.8$) is similar to that of other
205: systems in which H$_2$ is detected (see Fig.~15
206: of Ledoux et al. 2003). Substantial star-formation activity must have taken
207: place in the past history of this DLA system. Note that the age of the Universe
208: at $z=4.22$ is 1.5~Gyr (for $\Omega_\Lambda =0.73$, $\Omega_{\rm m}=0.27$ and
209: H$_0=71$ km s$^{-1}$ Mpc$^{-1}$).
210:
211: Nitrogen is thought to have both primary and secondary origins depending on
212: whether the seed carbon and oxygen nuclei are produced by the star itself or
213: are already present in the ISM from which the star formed. In the case
214: of secondary nitrogen production, the ratio of nitrogen to oxygen
215: abundances increases with increasing oxygen abundance, whereas for
216: primary production this ratio remains constant and nitrogen tracks oxygen
217: (this is the so-called primary plateau). In
218: the low-metallicity H\,{\sc ii} regions of nearby galaxies, [N/O] abundance
219: ratios tend to lie around the primary plateau (with [N/O$]=-0.73\pm 0.13$) for
220: [O/H$]<-1$, and have a secondary behaviour for [O/H$]>-1$
221: (van Zee et al. 1998; Izotov \& Thuan 1999). Similarly, most DLA systems show a
222: plateau at [N/O$]\approx -0.9$ (see
223: Centuri\'on et al. 2003; Pettini et al. 2002). The abundance of nitrogen
224: relative to sulfur at $z_{\rm abs}=4.224$ toward
225: PSS J\,1443$+$2724, [N/S$]=-0.75$, corresponds exactly to the primary nitrogen
226: production plateau (assuming [O/S$]=0$). If we assume the widely accepted
227: hypothesis that primary nitrogen is produced in intermediate-mass stars
228: ($4\leq M_\odot\leq 8$) during the Asymptotic Giant Branch phase
229: (Henry et al. 2000), then star-formation activity must have started in
230: this system at least $2-5\times 10^{8}$~yrs before $z=4.224$ accounting for the
231: characteristic lag time for intermediate-mass stars to eject nitrogen. This is
232: similar to what is observed in the $z_{\rm abs}=4.383$ system toward
233: BRI\,1202$-$0725 (D'Odorico \& Molaro 2004).
234:
235: Argon can be ionised in H\,{\sc i} regions and is therefore a tracer of
236: the radiation field. In the presently studied system, we find
237: [Ar/S$]=-0.30$, which is higher than in most other DLA systems
238: (Vladilo et al. 2003). This is consistent with the associated galaxy being in a
239: quiescent state at the time of observation, as also inferred below from
240: the analysis of H$_2$ lines.
241:
242: \subsection{Molecular hydrogen}
243:
244: The H$_2$ transitions are numerous and the spectral resolution of our data is
245: high enough to allow unambiguous detection and detailed analysis of H$_2$
246: (see Fig.~\ref{Fig1}). Great care has been exercised when fitting
247: the absorption lines with multi-component Voigt profiles. H$_2$ is detected
248: in four different rotational levels (${\rm J}=0$, 1, 2 and 3) of
249: the vibrational ground-state in three velocity components with total
250: column densities of $\log N({\rm H}_2)=17.67$, 17.97, 17.48 and 17.26
251: respectively (see Tables~\ref{Tabmetal} and \ref{TabH2}). The ${\rm J}=4$ level
252: is tentatively detected in the strongest component with
253: $\log N({\rm H}_2)\sim 14$. Blending is heavy however and this could be an
254: upper limit. The detection of H$_2$ confirms the previous claim that most of
255: the gas in this DLA system is cold (Howk et al. 2005). The total H$_2$
256: column density is one of the largest observed in DLA systems and the molecular
257: fraction is $\log f=-2.38\pm 0.13$,
258: with $f=2N({\rm H}_2)/(2N({\rm H}_2)+N($H\,{\sc i}$))$. HD is not detected
259: down to $3\sigma$ limits of $\log N($HD$)<13.6$ and 13.7 for ${\rm J}=0$ and 1
260: respectively, implying that $[$HD$/$H$_2]<-4.2$.
261:
262: The excitation temperatures $T_{01}$ for the two main components of the system
263: are 96 and 136~K respectively (see Table~\ref{TabH2}). The
264: excitation temperatures for higher rotational levels are slightly
265: larger, possibly as a consequence of the levels being populated by radiative
266: or formation pumping. By equating H$_2$ formation and destruction rates, we
267: write:
268: %
269: \begin{equation}
270: SIn({\rm H}_2) = Rn({\rm H})n
271: \end{equation}
272: %
273: where $n=n({\rm H})+2n({\rm H}_2)$ is the particle density, $I$ is the H$_2$
274: dissociation rate in the unshielded UV background (in s$^{-1}$) and
275: $S=[N({\rm H}_2)/10^{14}\ {\rm cm}^{-2}]^{-0.75}$ is the correction factor for
276: shielding (Draine \& Bertoldi 1996; Hirashita \& Ferrara 2005).
277: $R=4.1\times 10^{-17}[T/10^2\ {\rm K}]^{0.5}(Z/Z_\odot )D$ cm$^3$ s$^{-1}$ is
278: the H$_2$ formation rate per unit volume and time, $Z/Z_\odot =0.23$ is the
279: metallicity relative to Solar and $D=0.7$ is the fraction of metals into dust.
280:
281: The H$_2$ ${\rm J}=4$ level is mostly populated by pumping of lower rotational
282: levels by absorption of UV photons from the ambient radiation field
283: and subsequent radiative cascade, and by direct formation of the molecule in
284: this excited state. Following the simple procedure introduced
285: by Jura (1975), we write the equation describing the equilibrium of the
286: ${\rm J}=4$ level population as:
287: %
288: \begin{equation}
289: 9.1p_{4,0}SIn({\rm H}_2,{\rm J}=0)+0.19Rn({\rm H})n~=~A(4\rightarrow 2)n({\rm H}_2,{\rm J}=4)
290: \end{equation}
291: %
292: where $p_{4,0}=0.26$ is the pumping efficiency into the ${\rm J}=4$ level from
293: the ${\rm J}=0$ level
294: and $A(4\rightarrow 2)=2.8\times 10^{-9}$ s$^{-1}$ denotes the spontaneous
295: transition probability from ${\rm J}=4$ to ${\rm J}=2$.
296:
297: Combining the above equations and using the total column densities given in
298: Table~\ref{Tabmetal}, we find $Rn=4\times 10^{-16}$ s$^{-1}$. This is
299: probably an upper limit given the possibility that $N({\rm J}=4)$ be an upper
300: limit. Assuming a temperature of $T=150$~K, we derive a density
301: of $n_{\rm H}\leq 50$ cm$^{-3}$. This value could be slightly larger in case
302: of any $\alpha$-element enhancement. The corresponding H$_2$ dissociation
303: rate is $I\leq 3\times 10^{-10}$ s$^{-1}$. This is of the order of or
304: smaller than the highly variable photo-dissociation rate observed in the ISM
305: of the Galaxy (e.g., Hirashita \& Ferrara 2005). This probably indicates,
306: as expected above, that there is little on-going star formation in the galaxy
307: associated with this DLA system. This is consistent with the fact that
308: no emitting counterpart has been found using the Lyman-break techniques down
309: to $L\sim L^\star_{\rm LBG}/4$ (Prochaska et al. 2002) or in Lyman-$\alpha$
310: emission down to $2\times 10^{-17}$ erg cm$^{-2}$ s$^{-1}$ from 2D-spectroscopy
311: at the Canada-France-Hawaii Telescope (Ledoux 1999).
312:
313: \subsection{The Cosmic Microwave Background temperature}
314:
315: Absorption lines from the ground-state and first fine-structure excited level
316: of C\,{\sc i} are detected (see Fig.~\ref{Fig2}) at redshifts
317: of $z_{\rm abs}=4.22379$ and 4.22413, slightly larger than those of the
318: molecular components. The Doppler parameters are larger as well indicating
319: that the C\,{\sc i} gas spans a slightly larger velocity range than
320: the H$_2$ gas (see also Ledoux et al. 2003). We measure
321: $[$C\,{\sc i}$^\star /$C\,{\sc i}$]=-0.09\pm 0.05$ and $-0.21\pm 0.04$ in each
322: of the two components respectively. This is surprisingly small at
323: high redshift. Indeed, the value expected for excitation by the
324: Cosmic Microwave Background Radiation (CMBR) field alone, with a temperature of
325: 14.2~K at $z=4.22$, is $-0.15$ (Silva \& Viegas 2002). In addition, part of the
326: excitation can be due to UV or IR pumping although this is probably negligible
327: here. In Fig.~\ref{Fig3}, we plot the expected values for
328: $[$C\,{\sc i}$^\star /$C\,{\sc i}$]$ as a function of temperature of the
329: cold gas. The first C\,{\sc i} component has a density
330: ($n_{\rm H}\le 25$ cm$^{-3}$) that is similar to what is measured in the H$_2$
331: gas and is thus probably predominantly associated with it. The
332: second C\,{\sc i} component probably has a lower density and the corresponding
333: $[$C\,{\sc i}$^\star /$C\,{\sc i}$]$ ratio is consistent with what is expected
334: for excitation by the CMBR field alone.
335:
336: \section{Conclusion}
337:
338: The direct detection of H$_2$ at high redshift is a unique tool for
339: studying the ISM in the remote Universe. We have detected molecular hydrogen
340: in one of the highest redshift DLA systems known to date ($z_{\rm abs}=4.224$).
341: We have shown that the gas is probably the left-over of
342: intense star-formation activity that happened at least
343: $2\times 10^8$~yrs before the time of observation and that the associated
344: object is in a quiescent phase. The UV radiation field is of the order of
345: or smaller than that in the Galaxy and the gas is cold ($T\sim 90-180$~K).
346: The $[$C\,{\sc i}$^\star /$C\,{\sc i}$]$ ratio in one of the components of the
347: system is consistent with what is expected for excitation by the CMBR field
348: at this redshift ($T=14.2$~K). This study demonstrates the importance of
349: this kind of observations for our understanding of how star formation proceeds
350: at high redshift. The prospect of using $\gamma$-ray burst afterglows as
351: background sources for similar studies is exciting and will boost this field
352: in a near future (Draine \& Hao 2002).
353:
354:
355: \acknowledgements{
356: This work is based on data collected under prog. 072.A-0346 of the
357: European Southern Observatory with UVES mounted on the 8.2~m Kueyen
358: telescope at the Paranal Observatory, Chile. We thank the referee,
359: Paolo Molaro, for fruitful comments. PP and RS acknowledge support from the
360: Indo-French Centre for the Promotion of Advanced Research (Centre Franco-Indien
361: pour la Promotion de la Recherche Avanc\'ee) under contract No. 3004-3.}
362:
363: \begin{thebibliography}{}
364:
365: \bibitem[Ballester et al.(2000)]{2000Msg...101...31B} Ballester, P., Modigliani, A., Boitquin, O., Cristiani, S., Hanuschik, R., Kaufer, A., \& Wolf, S. 2000, The Messenger, 101, 31
366: %\bibitem[Berger et al.(2005)]{2005astroph0511498B} Berger, E., Penprase, B.~E., Cenko, S.~B., Kulkarni, S.~R., Fox, D.~B., Steidel, C.~C., \& Reddy, N.~A. 2005, \apj, submitted, astro-ph/0511498
367: \bibitem[Centuri\'on et al.(2003)]{2003A&A...403...55C} Centuri\'on, M., Molaro, P., Vladilo, G., P\'eroux, C., Levshakov, S.~A., \& D'Odorico, V. 2003, \aap, 403, 55
368: \bibitem[Cui et al.(2005)]{2005ApJ...633..649C} Cui, J., Bechtold, J., Ge, J., \& Meyer, D.~M. 2005, \apj, 633, 649
369: \bibitem[Dekker et al.(2000)]{2000SPIE.4008..534D} Dekker, H., D'Odorico, S., Kaufer, A., Delabre, B., \& Kotzlowski, H. 2000, \procspie, 4008, 534
370: \bibitem[D'Odorico \& Molaro(2004)]{2004A&A...415..879D} D'Odorico, V., \& Molaro, P. 2004, \aap, 415, 879
371: \bibitem[Draine \& Bertoldi(1996)]{1996ApJ...468..269D} Draine, B.~T., \& Bertoldi, F. 1996, \apj, 468, 269
372: \bibitem[Draine \& Hao(2002)]{2002ApJ...569..780D} Draine, B.~T., \& Hao, L. 2002, \apj, 569, 780
373: \bibitem[Ge \& Bechtold(1999)]{1999ASPC..156..121G} Ge, J., \& Bechtold, J. 1999, in Highly Redshifted Radio Lines, Carilli, C.~L., et al., eds., ASP Conf.~Ser. 156, 121
374: \bibitem[Grevesse \& Sauval(2002)]{2002ASR....30....3G} Grevesse, N., \& Sauval, A.~J. 2002, Adv. Space Res., 30, 3
375: \bibitem[Henry et al.(2000)]{2000ApJ...541..660H} Henry, R.~B.~C., Edmunds, M.~G., \& K\"oppen, J. 2000, \apj, 541, 660
376: \bibitem[Hirashita \& Ferrara(2005)]{2005MNRAS.356.1529H} Hirashita, H., \& Ferrara, A. 2005, \mnras, 356, 1529
377: \bibitem[Howk et al.(2005)]{2005ApJ...622L..81H} Howk, J.~C., Wolfe, A.~M., \& Prochaska, J.~X. 2005, \apjl, 622, L81
378: \bibitem[Izotov \& Thuan(1999)]{1999ApJ...511..639I} Izotov, Y.~I., \& Thuan, T.~X. 1999, \apj, 511, 639
379: \bibitem[Jura(1975)]{1975ApJ...197..581J} Jura, M. 1975, \apj, 197, 581
380: \bibitem[Ledoux(1999)]{1999PhDT........14L} Ledoux, C. 1999, Ph.D.~Thesis
381: \bibitem[Ledoux et al.(2002)]{2002A&A...392..781L} Ledoux, C., Srianand, R., \& Petitjean, P. 2002, \aap, 392, 781
382: \bibitem[Ledoux et al.(2003)]{2003MNRAS.346..209L} Ledoux, C., Petitjean, P., \& Srianand, R. 2003, \mnras, 346, 209
383: \bibitem[Levshakov \& Varshalovich(1985)]{1985MNRAS.212..517L} Levshakov, S.~A., \& Varshalovich, D.~A. 1985, MNRAS, 212, 517
384: \bibitem[Levshakov et al.(2002)]{2002ApJ...565..696L} Levshakov, S.~A., Dessauges-Zavadsky, M., D'Odorico, S., \& Molaro, P. 2002, \apj, 565, 696
385: \bibitem[Omont et al.(1996)]{1996Natur.382..428O} Omont, A., Petitjean, P., Guilloteau, S., McMahon, R.~G., Solomon, P.~M., \& P\'econtal, E. 1996, \nat, 382, 428
386: \bibitem[Petitjean et al.(2000)]{2000A&A...364L..26P} Petitjean, P., Srianand, R., \& Ledoux, C. 2000, \aap, 364, L26
387: \bibitem[Pettini et al.(2002)]{2002A&A...391...21P} Pettini, M., Ellison, S.~L., Bergeron, J., \& Petitjean, P. 2002, \aap, 391, 21
388: \bibitem[Prochaska et al.(2001)]{2001ApJS..137...21P} Prochaska, J.~X., et al. 2001, \apjs, 137, 21
389: \bibitem[Prochaska et al.(2002)]{2002AJ....123.2206P} Prochaska, J.~X., Gawiser, E., Wolfe, A.~M., Quirrenbach, A., Lanzetta, K.~M., Chen, H.-W., Cooke, J., \& Yahata, N. 2002, \aj, 123, 2206
390: \bibitem[Prochaska et al.(2003)]{2003ApJ...595L...9P} Prochaska, J.~X., Gawiser, E., Wolfe, A.~M., Castro, S., \& Djorgovski, S.~G. 2003, \apjl, 595, L9
391: \bibitem[Silva \& Viegas(2002)]{2002MNRAS.329..135S} Silva, A.~I., \& Viegas, S.~M. 2002, \mnras, 329, 135
392: \bibitem[Solomon \& Vanden Bout(2005)]{2005ARA&A..43..677S} Solomon, P.~M., \& Vanden Bout, P.~A. 2005, \araa, 43, 677
393: \bibitem[Vladilo et al.(2003)]{2003A&A...402..487V} Vladilo, G., Centuri\'on, M., D'Odorico, V., \& P\'eroux, C. 2003, \aap, 402, 487
394: \bibitem[van Zee et al.(1998)]{1998AJ....116.2805V} van Zee, L., Salzer, J.~J., Haynes, M.~P., O'Donoghue, A.~A., \& Balonek, T.~J. 1998, \aj, 116, 2805
395:
396: \end{thebibliography}
397:
398: \clearpage
399: %
400: \begin{table}
401: \caption {Total column densities of different species in the DLA~system
402: at $z_{\rm abs}=4.224$ toward PSS J\,1443$+$2724}
403: \begin{tabular}{llcc}
404: \hline
405: \hline
406: Species & Transition & $\log N\pm\sigma _{\log N}$ & Metallicity \\
407: & lines & & relative to Solar $^{\rm a}$\\
408: \hline
409: H\,{\sc i} & 1025,1215 & $20.95\pm 0.10$ & ... \\
410: H$_2$ & J=0 lines & $17.67\pm 0.06$ & ... \\
411: H$_2$ & J=1 lines & $17.97\pm 0.05$ & ... \\
412: H$_2$ & J=2 lines & $17.48\pm 0.12$ & ... \\
413: H$_2$ & J=3 lines & $17.26\pm 0.17$ & ... \\
414: H$_2$ & J=4 lines & $14.02\pm 0.14$ & ... \\
415: C\,{\sc i} & 1277,1280 & $13.12\pm 0.02$ & ... \\
416: C\,{\sc i}$^\star$ & 1277,1279 & $12.97\pm 0.03$ & ... \\
417: C\,{\sc i}$^{\star\star}$& 1277,1280 & $\le 12.5$ & ... \\
418: N\,{\sc i} & 953.4,1134.1& $15.52\pm 0.01$ & $-$1.38$\pm$0.10\\
419: Mg\,{\sc ii} & 1239,1240 & $15.98\pm 0.01$ & $-$0.55$\pm$0.10\\
420: S\,{\sc ii} & 1250 & $15.52\pm 0.01$ & $-$0.63$\pm$0.10\\
421: Ar\,{\sc i} & 1066 & $14.42\pm 0.01$ & $-$0.93$\pm$0.10\\
422: Fe\,{\sc ii} & 1081 & $15.33\pm 0.03$ & $-$1.12$\pm$0.10\\
423: \hline
424: \end{tabular}
425: \label{Tabmetal}
426: \flushleft
427: $^{\rm a}$ Solar abundances adopted from Grevesse \& Sauval (2002).\\
428: Note: errors in the column densities of metals correspond to the rms
429: errors from fitting the Voigt profiles. They do not include the uncertainties
430: related to the continuum placement.
431: \end{table}
432:
433: \clearpage
434: %
435: \begin{table}
436: \caption {Column densities and excitation temperatures from Voigt-profile
437: fitting of H$_2$ molecular lines}
438: \begin{tabular}{llcccc}
439: \hline
440: \hline
441: $z_{\rm abs}$ & Level & $\log N$ & $b\pm\sigma _b$ & $T_{\rm ex}$ & ${\rm J}_{\rm ex}$\\
442: & & $\pm\sigma _{\log N}$ & [km s$^{-1}$] & [K] & \\
443: \hline
444: 4.22371 & J=0 & $17.41^{+0.02}_{-0.05}$ & $0.8\pm 0.3$ & ... & ...\\
445: & J=1 & $17.59^{+0.03}_{-0.03}$ & \phantom{uu..}'' & 90-107 & 0-1\\
446: & J=2 & $17.15^{+0.05}_{-0.13}$ & \phantom{uu..}'' & 200-259 & 0-2\\
447: & J=3 & $16.53^{+0.17}_{-0.91}$ & \phantom{uu..}'' & 156-302 & 1-3\\
448: & J=4 & $<13.6$ $^{\rm a}$ & \phantom{uu..}'' & ... & ...\\
449: 4.22401 & J=0 & $17.34^{+0.09}_{-0.07}$ & $1.1\pm 0.3$ & ... & ...\\
450: & J=1 & $17.75^{+0.06}_{-0.08}$ & \phantom{uu..}'' & 104-179 & 0-1\\
451: & J=2 & $17.25^{+0.13}_{-0.14}$ & \phantom{uu..}'' & 218-377 & 0-2\\
452: & J=3 & $17.21^{+0.13}_{-0.13}$ & \phantom{uu..}'' & 337-530 & 1-3\\
453: & J=4 & $14.02^{+0.15}_{-0.07}$ & \phantom{uu..}'' & ... & ...\\
454: 4.22416 & J=0 & $14.34^{+0.00}_{-0.12}$ & $1.7\pm 0.5$ & ... & ...\\
455: & J=1 & $14.86^{+0.04}_{-0.13}$ & \phantom{uu..}'' & ... & ...\\
456: & J=2 & $14.61^{+0.06}_{-0.15}$ & \phantom{uu..}'' & ... & ...\\
457: & J=3 & $14.18^{+0.00}_{-0.16}$ & \phantom{uu..}'' & ... & ...\\
458: & J=4 & $<13.6$ $^{\rm a}$ & \phantom{uu..}'' & ... & ...\\
459: \hline
460: \end{tabular}
461: \label{TabH2}
462: \flushleft
463: $^{\rm a}$ $3\sigma$ detection limit.\\
464: Note: errors in the column densities correspond to the range
465: $b\pm\sigma_b$ of Doppler parameters and not to the rms errors from fitting the
466: Voigt profiles.
467: \end{table}
468:
469: \clearpage
470: %
471: \begin{figure}
472: \centering
473: \vbox{
474: \centering
475: \includegraphics[width=8.5cm,bb=60 320 403 769,clip,angle=0]{f1a.eps}
476: \includegraphics[width=8.5cm,bb=60 320 403 769,clip,angle=0]{f1b.eps}}
477: \caption[]{Velocity profiles of selected transition lines from the ${\rm J}=0$,
478: 1, 2, 3 and 4 rotational levels of the vibrational ground-state Lyman and
479: Werner bands of H$_2$ at $z_{\rm abs}=4.224$ toward PSS J\,1443$+$2724. The
480: best Voigt-profile fit is superposed on the observed spectrum with vertical
481: dotted lines marking the location of the three detected components.
482: %The origin of the velocity scale is set to $z_{\rm abs}=4.22401$.
483: }
484: \label{Fig1}
485: \end{figure}
486:
487: \clearpage
488: %
489: \begin{figure}
490: \centering
491: \includegraphics[width=8.5cm,bb=59 566 394 769,clip,angle=0]{f2.eps}
492: \caption[]{Velocity profiles of C\,{\sc i} and C\,{\sc i}$^\star$ transition
493: lines at $z_{\rm abs}=4.22379$ and 4.22413 toward PSS J\,1443$+$2724. The
494: best Voigt-profile fit is superposed on the observed spectrum with vertical
495: dotted lines marking the location of the two detected components of the
496: system and a third, possible component.}
497: \label{Fig2}
498: \end{figure}
499:
500: \clearpage
501: %
502: \begin{figure}
503: \centering
504: \includegraphics[width=8.5cm,bb=18 162 577 695,clip,angle=0]{f3.eps}
505: \caption[]{$[$C\,{\sc i}$^\star /$C\,{\sc i}$]$ ratio versus temperature.
506: Filled circles with error bars are the measurements in the
507: $z_{\rm abs}=4.22379$ and 4.22413 components toward PSS J\,1443$+$2724.
508: LTE predictions for two different values of the particle density are shown
509: by solid lines. The value expected for excitation by the CMBR field alone
510: is shown by the horizontal long-dashed line.}
511: \label{Fig3}
512: \end{figure}
513:
514: \end{document}
515: