1: \documentclass[12pt,preprint]{aastex}
2:
3: %\usepackage{epsfig,emulateapj5,apjfonts}
4: %\usepackage{epsfig}
5: %%%%%%%%%%%%%%%%%%%%% macro definitions %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
6: \newcommand \Angstrom {\,{\rm \AA}}
7: \newcommand \AU {\,{\rm AU}}
8: \newcommand \cm {\,{\rm cm}}
9: \newcommand \mm {\,{\rm mm}}
10: \newcommand \erg {\,{\rm erg}}
11: \newcommand \eV {\,{\rm eV}}
12: \newcommand \g {\,{\rm g}}
13: \newcommand \K {\,{\rm K}}
14: \newcommand \pc {\,{\rm pc}}
15: \newcommand \kpc {\,{\rm kpc}}
16: \newcommand \s {\,{\rm s}}
17: \newcommand \sr {\,{\rm sr}}
18: \newcommand \yr {\,{\rm yr}}
19: \newcommand \yrs {\,{\rm yrs}}
20: \newcommand \Myr {\,{\rm Myr}}
21: \newcommand \Myrs {\,{\rm Myrs}}
22: \newcommand \Gyr {\,{\rm Gyr}}
23: \newcommand \nH {n_{\rm H}}
24: \newcommand \simlt {\lesssim}
25: \newcommand \simgt {\gtrsim}
26: \newcommand \gtsim {\gtrsim}
27: \newcommand \ltsim {\lesssim}
28: \newcommand \um {\mu{\rm m}}
29: \newcommand \mum {\,{\rm \mu m}}
30: \newcommand \ppm {\,{\rm ppm}}
31: \newcommand \Teff {T_{\rm eff}}
32: \newcommand \Msun {\,{M_\odot}}
33: \newcommand \Lsun {\,{L_\odot}}
34: \newcommand \mH {m_{\rm H}}
35: \newcommand \NH {N_{\rm H}}
36: \newcommand \simali {\sim\,}
37: %
38: \newcommand{\figwidth}{4.0in}
39: %\newcommand{\figwidth}{6.0in}
40: \newcommand{\btdnote}[1]{{\bf[#1]}}
41: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
42: %------- delete following for submission to ApJ --------
43: %\pagestyle{myheadings}
44: % today's date
45: %use number register 200 for "decade"
46: %\countdef\decade=200
47: %\decade=0
48: %\advance\decade by \year
49: %\advance\decade by -2000 %to suppress two leading digits of yearb
50: %\countdef\hours=201
51: %\hours=0
52: %\advance\hours by \time
53: %\divide\hours by 60
54: %\countdef\mins=202
55: %\mins=0
56: %\advance\mins by \hours
57: %\multiply\mins by 60
58: %\multiply\hours by 100
59: %\countdef\miltime=203
60: %\miltime=0
61: %\advance\miltime by \hours
62: %\advance\miltime by \time
63: %\advance\miltime by -\mins
64: %\def\today{\number\decade.\number\month.\number\day.\number\miltime}
65: %\markright{\today: DRAFT}
66:
67: \shorttitle{Dust Extinction of GRB Host Galaxies}
68:
69: %\received{2003 September 18}
70: \begin{document}
71: \title{
72: %------------- enable for labelling preprint ---------------------------
73: % \vspace*{-2.0em}
74: % {\normalsize\rm submitted to {\it The Astrophysical Journal Letters}}\\
75: % \vspace*{1.0em}
76: Dust Extinction of Gamma-ray Burst Host Galaxies:
77: Identification of Two Classes?
78: %\\{\small DRAFT: \today ~~}
79: }
80: \author{S.L. Chen\altaffilmark{1},
81: Aigen Li\altaffilmark{1,2},
82: and D.M. Wei\altaffilmark{1,3}}
83:
84: \altaffiltext{1}{Purple Mountain Observatory, Chinese Academy of
85: Sciences, Nanjing 210008, China; {\sf fugue@pmo.ac.cn}}
86: %\altaffiltext{2}{National Astronomical Observatories, Chinese
87: % Academy of Sciences, Beijing, 100012, China}
88: \altaffiltext{2}{Department of Physics and Astronomy, University of
89: Missouri, Columbia, MO 65211; {\sf lia@missouri.edu}}
90: \altaffiltext{3}{Joint Center for Particle Nuclear Physics and
91: Cosmology of Purple Mountain Observatory --
92: Nanjing University, Nanjing 210008, China;
93: {\sf dmwei@pmo.ac.cn}}
94:
95: \begin{abstract}
96: Dust in the host galaxies of gamma-ray bursts (GRBs)
97: dims and reddens their afterglow spectra.
98: Knowledge of the nature of this dust
99: is crucial for correcting for extinction,
100: providing clues to the nature of GRB progenitors,
101: and probing the interstellar medium of high-redshift galaxies
102: as well as the nature of cosmic dust when the universe was
103: much younger and galaxies were much less evolved.
104: The dust and extinction properties of GRB host galaxies
105: are still poorly known. Unlike previous work, we derive
106: in this {\it Letter} the extinction curves for 10 GRB host
107: galaxies without a priori assumption of any specific
108: extinction types (such as that of the Milky Way,
109: or the Small/Large Magellanic Clouds).
110: It is found that there appears to exist two different types
111: of extinction curves: one is relatively flat and gray,
112: the other displays a steeper dependence on inverse
113: wavelength, closely resembling that of the Milky Way but
114: with the 2175$\Angstrom$ feature removed.
115: %We model these two different types of extinction
116: %in terms of the silicate-graphite interstellar grain
117: %model. Their implications are also briefly discussed.
118: \end{abstract}
119:
120: \keywords{dust, extinction --- gamma rays: bursts}
121:
122: \section{Introduction}
123: The existence of dust extinction towards gamma-ray bursts (GRBs)
124: and their afterglows has been well established
125: {\it observationally} through
126: (1) ``dark bursts'' -- nearly $\simali$60\% of the X-ray afterglows
127: have no optical counterparts -- the undetected optical afterglows
128: may have been extinguished by dust in the host galaxy
129: %(see Lazzati, Covino, \& Ghisellini 2002, and references therein);
130: (see Lazzati et al.\ 2002 and references therein);
131: (2) reddening -- the optical/near-infrared (IR) spectral energy
132: distributions (SEDs) of afterglows deviate from that expected
133: from standard models,\footnote{%
134: As any host-galaxy extinction would result
135: in a steepening of the intrinsic spectral slope,
136: a steeper (observed) slope (than all intrinsic spectral
137: slopes) betrays the presence of host galaxy dust.
138: }
139: indicative of dust reddening
140: %(see Kann, Klose, \& Zeh 2006, and references therein);
141: (see Kann et al.\ 2006 and references therein);
142: (3) the reduced Balmer line ratios
143: (as a consequence of dust extinction)
144: in the spectra of some GRB host galaxies
145: (e.g. see Djorgovski et al.\ 1998)
146: compared with the expected ratios for
147: the standard Case B recombination (Osterbrock \& Ferland 2006);
148: (4) the depletion of dust-forming heavy elements
149: such as Si and Fe in some host galaxies
150: %(e.g. see Savaglio, Fall, \& Fiore 2003);
151: (e.g. see Savaglio et al.\ 2003);
152: and {\it theoretically} (5) through the association of
153: (at least long-duration [$\simgt$\,2\,s]) GRBs
154: with massive stars and star-forming regions
155: embedded in dense clouds of dust and gas (Paczy\'nski 1998).
156:
157: A precise knowledge of the nature (e.g. size, composition)
158: of the dust in GRB host galaxies is very useful for
159: (1) correcting for the extinction of afterglows
160: from X-ray to near-IR wavelengths;
161: (2) constraining the nature of the GRB progenitors
162: (i.e. collapsing massive stars or merging neutron binaries);
163: and (3) understanding the interstellar medium (ISM) of
164: high-redshift galaxies and the cosmic star formation history
165: %(e.g. see Ramirez-Ruiz, Trentham, \& Blain 2002).
166: (e.g. see Ramirez-Ruiz et al.\ 2002).
167:
168: However, our knowledge of the dust in GRB host galaxies
169: is very limited. Previous studies all are limited to a qualitative
170: and rather speculative analysis of the extinction curves derived by
171: fitting the observed optical/near-IR SEDs of GRB afterglows with a
172: power-law (approximating their intrinsic spectra) reddened by an
173: extinction law of known types (mostly either that of the Milky Way
174: [MW], the Small Magellanic Cloud [SMC], or that of the Large
175: Magellanic Cloud [LMC]) adopted as a priori (e.g. see Stratta et
176: al.\ 2004, Kann et al.\ 2006).
177:
178: In this {\it Letter}, by assuming that the GRB afterglows
179: have a power-law intrinsic spectrum as expected from
180: the standard fireball model, we obtain the extinction curves
181: for 10 GRB host galaxies without a priori assumption of
182: the extinction law. The size and composition of the dust
183: are modeled quantitatively in terms of the silicate-graphite
184: interstellar grain model.
185: %The implications for the environments of GRB progenitors
186: %are also discussed.
187:
188: \section{Method}
189: In the standard fireball model, the temporal ($t$)
190: and frequency ($\nu$) dependence of GRB afterglows
191: can be well described by
192: $F_\nu(t)\propto t^{-\alpha} \nu^{-\beta}$,
193: where both $\alpha$ and $\beta$ are related with
194: the electron energy distribution index $p$
195: ($dn\propto E^{-p}\,dE$ for the energy distribution
196: of the shock-heated electrons; see Sari et al.\ 1998).
197: Therefore, with the decay index $\alpha$ determined,
198: $p$ as well as the intrinsic spectral index $\beta$
199: can be well constrained.
200:
201: In addition to the extinction of the host galaxy,
202: GRB afterglows are also subject to the extinction
203: of our own MW galaxy. We correct for the latter
204: using the reddening maps of
205: %Schlegel, Finkbeiner, \& Davis (1998).
206: Schlegel et al.\ (1998).
207: The host's spectrum is redshifted to longer wavelengths
208: where any errors in the Schlegel et al.\ map are not important.
209:
210: Let $F_{\lambda}^{0}$ be the intrinsic spectrum. With the Galactic
211: reddening corrected, the observed wavelength ($\lambda$)-dependent
212: spectrum is
213: $F_\lambda\,=\,F_{\lambda}^{0}\,\exp\left(-A_\lambda\right)$, where
214: $A_\lambda$ is the extinction from the host galaxy. Setting $V$ band
215: as the zero point, we obtain
216: $F_\lambda/F_V\,=\,\left(\lambda_V/\lambda\right)^{2-\beta_0}
217: \exp\left(A_V-A_\lambda\right)$, where $\lambda$ and
218: $\lambda_V$\,$\equiv$\,5500$\Angstrom$ are both measured in the rest
219: frame. With $\beta_{0}$ estimated from the decay index $\alpha$, we
220: can obtain $A_\lambda - A_V$ from $F_\lambda/F_V$. To determine
221: $A_V$, we fit a simple function $A_\lambda - A_V=
222: a/\left[1+\left(\lambda/\lambda_B\right)^2\right]+b$ for $\lambda >
223: \lambda_V$ and then extrapolate this function to $\lambda\rightarrow
224: \infty$ to get $A_V$ (=$-b$) since $A_\lambda\rightarrow 0$ as
225: $\lambda\rightarrow \infty$.
226: %(e.g. see Kr\"ugel 2003).
227:
228: The standard silicate-graphite interstellar dust model,
229: consisting of a mixture of spherical silicate and graphite grains,
230: is shown successful in reproducing the extinction and IR
231: emission of the MW galaxy, SMC and LMC
232: (Weingartner \& Draine 2001; Li \& Draine 2001, 2002).
233: %
234: We will apply this grain-mixture to model the extinction curves
235: $A_\lambda/A_V$ determined for the GRB host galaxies, but with
236: a {\it simpler} functional formula
237: for the dust size distribution:
238: %(than that of Weingartner \& Draine 2001):
239: $dn\sim a^{-\eta}\,\exp(-a/a_c)\,da$
240: for both grain types, where $a$ is the grain radius,
241: ranging from 0.05 to 2.5$\mum$,
242: %(we take all grains to be spherical),
243: $\eta$ is the power-law index,
244: $a_c$ is the cut-off size, and $dn$ is the number of grains
245: in the size interval [$a$,\,$a+da$].
246: The mass fraction of graphite dust is $f_{\rm gra}$
247: [for silicates it is $(1-f_{\rm gra})$].
248: In Figure 1 we show that the MW, SMC and LMC
249: extinction curves are well fit by the silicate-graphite
250: model with this simple dust size distribution function.
251:
252: \section{Data}
253: To construct an extinction curve, one requires simultaneous
254: multi-band photometry. We carefully collected such data from
255: literature for ten bursts taken when their decays were in a steady
256: power-law state(e.g. see Panaitescu \& Kumar 2001, Fan \& Piran 2006
257: for very detailed analysis). The optical and near-IR UBVRIJHK fluxes
258: $F_\nu$ of these bursts are tabulated in Table 1. Also tabulated is
259: $\beta_0$, the intrinsic spectrum index calculated from fitting the
260: decay index based on the standard model of afterglow (see \S2).
261:
262: To compare the extinction properties in the rest frame
263: of each burst, in the observer frame we need to calculate
264: $A_\lambda - A_{[V(1+z)]}$.
265: The results are shown in Table 2.
266: We correct the wavelength with a factor $(1+z)$ to
267: get $A_\lambda - A_V$ in the rest frame.
268: Also shown in Table 2 are $A_V$, obtained by extrapolating
269: $A_\lambda - A_V$ to $\lambda \rightarrow \infty$
270: (see \S2).
271:
272: \section{Extinction Curves: Two Different Types?}
273: The extinction curves (normalized to the $V$ band)
274: derived for these 10 host galaxies are shown in
275: Figures 2 and 3. A common feature for these curves
276: is the absence of the 2175$\Angstrom$ bump,
277: which is the strongest absorption band
278: in the MW extinction (Li 2005).
279: They appear to fall into two categories:
280: one is relatively gray, showing a much flatter increase
281: with inverse wavelength $\lambda^{-1}$
282: (we call it ``Type-I''; see Fig.\,2),
283: the other displays a much steeper,
284: almost linear increase with $\lambda^{-1}$
285: (we call it ``Type-II''; see Fig.\,3).
286:
287:
288: In view of its possible practical use (e.g. in correcting for
289: extinction of afterglows), following Fitzpatrick \& Massa (1990),
290: we take an analytical fit to these 2 extinction types:
291: \begin{equation}
292: A_\lambda/A_V =
293: \left\{\begin{array}{ll}
294: c_1 + c_2\,x + c_3\,D(x;\gamma,x_0) + c_4 F(x), & x\ge 1\mum^{-1}, \\
295: k\,x^{1.84}, & x\,<\,1\mum^{-1},\\
296: \end{array} \right .
297: \end{equation}
298: where $x\equiv \lambda^{-1}$,
299: $D(x;\gamma,x_0)\,=\,x^2/\left[\left(x^2-x^2_0\right)^2
300: + x^2\gamma^2\right]$, and
301: \begin{equation}
302: F(x) =
303: \left\{\begin{array}{ll}
304: 0.5392\,\left(x-5.9\right)^2 + 0.05644\,\left(x-5.9\right)^3,
305: & x\ge 5.9\mum^{-1},\\
306: 0, & x\,<\,5.9\mum^{-1}.\\
307: \end{array} \right .
308: \end{equation}
309: In this fit, $c_1$ and $c_2$ determine a linear ``background''
310: term; $c_3$ determines the strength of the 2175$\Angstrom$
311: feature which is represented by the ``Drude profile'' term
312: $D(x;\gamma,x_0)$ -- the theoretical profile
313: for a classic damped harmonic oscillator
314: (Bohren \& Huffman 1983);
315: $\gamma$ and $x_0$ are respectively
316: the FWHM and peak position of the Drude profile;\footnote{%
317: In the Galactic ISM, the strength and width of
318: the 2175$\Angstrom$ extinction bump vary with
319: environment while its peak position is quite invariant
320: (see Li 2005).
321: %the central wavelength of this feature varies by only
322: %$\pm$0.46\% (2$\sigma$) around 2175$\Angstrom$
323: %(see Whittet 2003).
324: We therefore fix $x_0$ at 4.6$\mum^{-1}$.
325: }
326: $c_4$ determines the far-UV curvature term represented
327: by $F(x)$.
328: The resulting analytical fits are plotted in Figures 2 and 3;
329: and the fitted parameters $c_1$, $c_2$, $c_3$, $c_4$,
330: and $\gamma$ are tabulated in Table 3.
331:
332:
333: To quantify the dust properties, we fit both extinction types
334: in terms of the silicate-graphite dust model.
335: The model fits and parameters are respectively
336: shown in Figures 2,3 and in Table 4.
337: The model for the flat ``Type-I'' extinction curve
338: is dominated by {\it large} silicate grains
339: (with $f_{\rm gra} \approx 0$ and a flat size distribution).
340: In contrast, the model for the steep ``Type-II'' extinction
341: has a much steeper size distribution, indicating the richness
342: of {\it small} dust in the host galaxies
343: with a ``Type-II'' extinction curve.
344:
345:
346: \section{Discussion}
347: Previous efforts in deriving the dust extinction of GRB host
348: galaxies all assume an initial power-law for the GRB intrinsic
349: spectrum (as expected from the standard fireball model), and an
350: extinction curve of known types such as that of MW, SMC, LMC,
351: %the Calzetti law (Calzetti, Kinney, \& Storchi-Bergmann 1994)
352: the Calzetti et al.\ (1994) law
353: suitable for local starburst galaxies,
354: %and the Maiolino law (Maiolino, Marconi, \& Oliva 2001)
355: and the Maiolino et al.\ (2001) law
356: suggested for the dust in the circumnuclear
357: region of AGNs (e.g. see Stratta et al.\ 2004 and Kann et al.\ 2006
358: for recent examples).
359: Our approach, without the need for an initial assumption
360: of specified extinction types, is more favourable because
361: of the lack of a priori knowledge of the extinction in the GRB
362: hosts. The fact that neither of the two extinction types derived in
363: this work resembles that of MW, SMC and LMC (see Figs.\,2,3)
364: challenges the initial assumption of specific extinction types.
365:
366: In consistent with previous studies, this work
367: also finds no evidence for the 2175$\Angstrom$
368: extinction bump.\footnote{%
369: The only exception is GRB 991216.
370: Its afterglow spectrum shows a depression
371: in flux in between 4000$\Angstrom$ and 5500$\Angstrom$,
372: suggesting the presence of a {\it red}
373: 2175$\Angstrom$-type extinction bump
374: (at $\simali$2360\AA) in the host of GRB 991216
375: (Vreeswijk et al.\ 2006).
376: Similar red 2175$\Angstrom$ bumps have been previously
377: seen in a few UV-strong, hydrogen-poor stars
378: %in the MW galaxy (Greenstein 1981, Buss et al. 1989).
379: in the MW galaxy (see Li 2005 and references therein).
380: }
381: The carrier of this bump remains
382: unidentified over 40 years after its first detection.
383: %(Stecher 1965).
384: It is generally believed to be caused by aromatic
385: carbonaceous (graphitic) materials,
386: likely a cosmic mixture of polycyclic aromatic
387: hydrocarbon (PAH) molecules
388: %(Joblin, L\'eger \& Martin 1992; Li \& Draine 2001).
389: (see Li 2005 and references therein).
390: The nondetection of the 2175$\Angstrom$ extinction bump
391: in the GRB hosts could be explained in terms of
392: (1) the depletion of its carrier (e.g. PAHs)
393: by condensation onto the ice mantles coated
394: on the dust in the dense star-forming regions
395: in which GRBs are embedded, and
396: (2) the destruction of its carrier by UV/X-ray
397: radiation in the immediate vicinities (up to
398: $\simali$20\pc) of bursts.
399:
400: Our Type-I extinction curve (see Fig.\,2) is more
401: gray (i.e. with a much weaker dependence on wavelength)
402: than that of MW, SMC, and LMC.
403: Gray dust has been invoked by a number of authors
404: (e.g. Savaglio et al.\ 2003, Savaglio \& Fall 2004;
405: Stratta et al.\ 2004, 2005)
406: to account for the small ratios of extinction
407: and/or reddening (derived from fitting the afterglow SEDs)
408: to H column densities (determined from X-ray or
409: Ly$\alpha$ absorption) of GRB hosts, which are
410: usually smaller than that of MW
411: by a factor of $\simali$10--100
412: (Galama \& Wijers 2001, {\v S}imon et al.\ 2001,
413: Hjorth et al.\ 2003, Vreeswijk et al.\ 2004).\footnote{%
414: Alternatively, these results can be explained by
415: an intrinsicaly low dust content --
416: a low metallicity and/or a low level of depletion
417: of the dust-forming heavy elements in the burst
418: environment (Fynbo et al.\ 2003, Hjorth et al.\ 2003,
419: Vreeswijk et al.\ 2004, Fruchter et al.\ 2006).
420: However, Savaglio et al.\ (2003) found that in
421: the host galaxies of 3 GRBs
422: both the column densities of metals (as indicated by Zn)
423: and the depletion of heavy elements (such as Fe, Si and Cr)
424: are large, indicating a large dust content.
425: }
426: Gray dust could be created by
427: (1) the preferential destruction of small grains
428: by the intense UV and X-ray radiation from the GRB
429: %(Waxman \& Draine 2000, Fruchter, Krolik, \& Rhoads 2001,
430: %Perna, Lazzati, \& Fiore 2003) in the immediate GRB
431: (Waxman \& Draine 2000, Fruchter et al.\ 2001,
432: Perna et al.\ 2003) in the immediate GRB
433: environment, say, within $\simali$10--20\,pc around the burster
434: (Hjorth et al.\ 2003, Savaglio et al.\ 2003, Stratta et al.\ 2005),
435: and (2) the growth of dust through coagulation
436: (which also leads to the depletion of small grains)
437: in the high-density environments such as the cores of
438: star-forming regions harboring GRB events
439: (Maiolino et al. 2001, Stratta et al.\ 2004).
440: Both mechanisms naturally lead to a dust size distribution
441: skewed toward large grains, producing an extinction curve
442: weakly dependent on wavelength.
443:
444: Nongray extinction has also been reported
445: (e.g. for the host of GRB 010222 by Galama et al.\ 2003).
446: Our nongray, Type-II extinction curve (see Fig.\,3)
447: shows a steep increase with $\lambda^{-1}$,
448: indicating that small grains are more abundant
449: in the GRB host galaxies with a Type-II
450: extinction than in those with a Type-I extinction.
451: This is probably caused by a less complete destruction of
452: small grains (e.g. associated with less energetic bursts
453: of less intense UV/X-ray radiation), and/or a slower
454: coagulational growth of dust (e.g. associated with less dense
455: environments or metal-poorer galaxies).
456: Since our sample is small, we do not want to
457: overinterpret its implications.
458: A more systematic investigation of the extinction
459: and dust properties of a large sample of
460: GRB host galaxies is in progress.
461:
462: Finally, if one really wants to make a priori assumption
463: of the extinction spectral shape when deriving the GRB host
464: galaxy dust extinction, we suggest the adoption of
465: the Fitzpatrick \& Massa (1990) formulae instead of
466: that of MW, SMC or LMC. By first visually inspecting
467: the GRB spectrum, one can set $c_3=0$ (in eq.[1])
468: if there is no sign of the 2175$\Angstrom$ extinction
469: bump or curvature on large wavelength intervals.
470:
471: \acknowledgments
472: We thank Y.Z. Fan, D.A. Kann, G. Stratta, and the anonymous
473: referee for helpful comments. S.L.C. and D.M.W. are supported
474: by the NSFC grants 10225314 and 10233010,
475: and the National 973 Project NKBRSF G19990754.
476: A.L. is supported by the University of Missouri
477: Summer Research Fellowship, the University of Missouri
478: Research Board, and NASA/Spitzer Theory Programs.
479:
480: \begin{thebibliography}{}
481: \bibitem[]{598}Bloom, J.~S., van Dokkum, P.~G., Bailyn, C.~D.,
482: Buxton, M.~M., Kulkarni, S.~R., \& Schmidt, B.~P.\
483: 2004, \aj, 127, 252
484: \bibitem[]{720}Bohren, C.F., \& Huffman, D.R. 1983,
485: Absorption and Scattering of Light by
486: Small Particles (New York: Wiley)
487: %\bibitem[]{601} Buss, R.~H., Snow, T.~P., \& Lamers, H.~J.~G.~L.~M.\
488: % 1989, \apj, 347, 977
489: \bibitem[]{603} Calzetti, D., Kinney, A.~L., \& Storchi-Bergmann, T.\
490: 1994, \apj, 429, 582
491: \bibitem[]{605} Covino, S., et al.\ 2003, A\&A, 404, L5
492: \bibitem[]{606} Djorgovski, S.~G., Kulkarni, S.~R., Bloom, J.~S.,
493: Goodrich, R., Frail, D.~A., Piro, L.,
494: \& Palazzi, E.\ 1998, \apjl, 508, L17
495: %\bibitem[]{609} Draine, B. T. 2003, ARA\&A, 41, 241
496: \bibitem[]{610} Fan, Y. Z., \& Piran, T. 2006, MNRAS, 369, 197
497: \bibitem[]{611} Fitzpatrick, E. L., \& Massa, D. L.\
498: 1990, ApJS, 72, 163
499: \bibitem[]{613} Frail, D. A., et al.\ 2003, ApJ, 590, 992
500: \bibitem[]{614} Fruchter, A., Krolik, J.~H., \& Rhoads, J.~E.\
501: 2001, \apj, 563, 597
502: \bibitem[]{616} Fruchter, A., et al.\ 2006, Nature, 461, 443
503: \bibitem[]{617} Fynbo, J.~P.~U., et al.\
504: 2003, \aap, 406, L63
505: \bibitem[]{619} Galama, T.~J., \& Wijers, R.~A.~M.~J.\ 2001,
506: \apjl, 549, L209
507: \bibitem[]{621} Galama, T.~J., et al.\ 2003, \apj, 587, 135
508: %\bibitem[]{622} Greenstein, J.~L.\ 1981, \apj, 245, 124
509: %\bibitem[]{623} Groot, P.~J., et al.\ 1998, \apjl, 493, L27
510: \bibitem[]{624} Hjorth, J., et al.\ 2003, \apj, 597, 699
511: \bibitem[]{625} Holland, S. T., et al.\ 2003, AJ, 125, 2291
512: \bibitem[]{626} Jakobsson, P., et al.\ 2003, A\&A, 408, 941
513: \bibitem[]{627} Jakobsson, P., et al.\ 2004, A\&A, 427,785
514: \bibitem[]{628} Jensen, B. L., et al.\ 2001, A\&A, 370, 909
515: %\bibitem[]{629} Joblin, C., L\'{e}ger, A., \& Martin, P.\
516: % 1992, ApJ, 393, L79
517: \bibitem[]{631} Kann, D.~A., Klose, S., \& Zeh, A.\
518: 2006, \apj, 641, 993
519: \bibitem[]{633} Klose, S., et al.\ 2004, AJ, 128,1942
520: %\bibitem[]{634} Kr\"ugel, E. 2003, Physics of Interstellar Dust
521: % (Bristol: IoP)
522: %\bibitem[]{636} Lazzati, D., Perna, R., \& Ghisellini, G.\
523: % 2001, \mnras, 325, L19
524: \bibitem[]{638} Lazzati, D., Covino, S., \& Ghisellini, G.\
525: 2002, \mnras, 330, 583
526: \bibitem[]{} Li, A.\ 2005, J. Phys.: Conf. Ser., 6, 229
527: (astro-ph/0503569)
528: \bibitem[]{640} Li, A., \& Draine, B. T. 2001, ApJ, 554,778
529: \bibitem[]{641} Li, A., \& Draine, B.T. 2002, ApJ, 576, 762
530: \bibitem[]{642} Maiolino, R., Marconi, A., \& Oliva, E.\
531: 2001, \aap, 365, 37
532: \bibitem[]{644} Osterbrock, D.~E., \& Ferland, G.~J.\ 2006,
533: Astrophysics of Gaseous Nebulae and Active
534: Galactic Nuclei
535: (2nd ed; Sausalito: Univ. Sci. Books)
536: \bibitem[]{648} Paczy\'{n}ski, B.\ 1998, \apjl, 494, L45
537: \bibitem[]{649} Panaitescu, A., \& Kumar, P. 2001, \apj, 554, 667
538: \bibitem[]{650} Perna, R., Lazzati, D., \& Fiore, F.\
539: 2003, \apj, 585, 775
540: \bibitem[]{652} Ramirez-Ruiz, E., Trentham, N., \& Blain, A.~W.\
541: 2002, \mnras, 329, 465
542: \bibitem[]{654} Sari, R., Piran, T., \& Narayan, R.\
543: 1998, \apjl, 497, L17
544: \bibitem[]{656} Savaglio, S., \& Fall, S.~M.\
545: 2004, \apj, 614, 293
546: \bibitem[]{658} Savaglio, S., Fall, S.~M., \& Fiore, F.\
547: 2003, \apj, 585, 638
548: \bibitem[]{660} {\v S}imon, V., Hudec, R., Pizzichini, G.,
549: \& Masetti, N.\ 2001, \aap, 377, 450
550: %\bibitem[]{662} Stecher, T.P.\ 1965, ApJ, 142, 1683
551: \bibitem[]{663} Stratta, G., Fiore, F., Antonelli, L.~A., Piro, L.,
552: \& De Pasquale, M.\ 2004, \apj, 608, 846
553: \bibitem[]{665} Stratta, G., Perna, R., Lazzati, D., Fiore, F.,
554: Antonelli, L.~A., \& Conciatore, M.~L.\ 2005, \aap, 441, 83
555: \bibitem[]{667} Schlegel, D. J., Finkbeiner, D. P., \& Davis, M.\
556: 1998, ApJ, 500, 525
557: \bibitem[]{669} Vreeswijk, P.~M., et al.\ 2004, \aap, 419, 927
558: \bibitem[]{670} Vreeswijk, P.~M., et al.\ 2006, \aap, 447, 145
559: \bibitem[]{671} Waxman, E., \& Draine, B.~T.\ 2000, \apj, 537, 796
560: \bibitem[]{672} Weingartner, J.C., \& Draine, B.T. 2001, ApJ, 548, 296
561: %\bibitem[]{673} Whittet, D.C.B. 2003, Dust in the Galactic Environment
562: % (2nd ed; Bristol: IoP)
563: \end{thebibliography}
564:
565: \clearpage
566:
567: \begin{figure}
568: \plotone{f1.eps}
569: \caption{
570: \label{fig:mwsmclmc}
571: Fitting the extinction curves
572: of the Milky Way galaxy, the SMC and LMC
573: by the silicate-graphite model with
574: a simple dust size distribution
575: $dn/da\propto a^{-\eta}\,\exp(-a/a_c)$.
576: Points -- observational data;
577: %(taken from Whittet 2003);
578: solid lines -- model fits.
579: }
580: \end{figure}
581:
582: \clearpage
583:
584: \begin{figure}
585: \plotone{f2.eps}
586: \caption{
587: \label{fig:typeI}
588: Flat, ``Type-I'' extinction curves
589: (normalized to the $V$ band) for 4 GRB host galaxies.
590: Both axes are in the rest-frame of the GRB host.
591: Also plotted are the model (black) and analytical (magenta)
592: fits to the observed extinction curves.
593: The MW, SMC and LMC extinction curves
594: are also shown for comparison.
595: }
596: \end{figure}
597:
598: \clearpage
599:
600: \begin{figure}
601: \plotone{f3.eps}
602: \caption{
603: \label{fig:typeII}
604: Same as Figure 2 but for the relatively steep,
605: ``Type-II'' extinction curves derived for 6 GRB
606: host galaxies.
607: }
608: \end{figure}
609:
610: \clearpage
611:
612: \begin{table}
613: \caption[]{Optical to near-IR fluxes of 10 GRB afterglows\label{tab:tbl1}}
614: {\footnotesize
615: \begin{tabular}{cccccccccccccc}
616: \tableline\tableline GRB & $z$ & $\beta_0$ & $A_V^{\rm MW}$
617: &\multicolumn{8}{c}{$F_\nu$ ($\mu Jy$)} & $t$ & References\\
618: \cline{5-12}
619: & & & (mag) & $U$ & $B$ & $V$ & $R$ & $I$ &$J$ &$H$ & $K$ & (days)\\
620: \cline{5-12} \tableline
621: 980703 & 0.966 & 1.07 &0.192 & \nodata &2.7 &3.4 &4.6 &7.6 &10.8 &21.8 &25.5 & 5 & 1 \\
622: 000301C & 2.0404 & 0.56 &0.308 &11.6 &14.1 &17.3 &21.6 &26.1 &35.3 &44.9 &60.6 &3.1 & 2 \\
623: 011211 & 2.14 & 0.63 &0.142 &11.4 &13.5 &15.5 &19.1 &21.9 &29.5 &\nodata &42.0 &0.57 & 3 \\
624: 020405 & 0.691 & 0.5 &0.180 &5.4 &7.2 &9.5 &10.9 &15.7 &28.6 &34.9 &42.5 &1.98 & 4 \\
625: 020813 & 1.255 & 0.52 &0.370 &6.1 &7.7 &9.0 &11.1 &13.7 &22.4 &33.2 &32.5 &1.93 & 5 \\
626: 021004 & 2.33 & 0.45 &0.198 &4.3 &7.2 &11.1 &13.7 &17.2 &\nodata &32.6 &\nodata &5.5 & 6 \\
627: 030226 & 1.986 & 0.33 &0.062 &\nodata &17.3 &20.4 &23.9 &27.6 &35.1 &42.1 &56.2 &0.8 & 7 \\
628: 030323 & 3.3718 & 0.28 &0.163 &\nodata &2.0 &5.2 &7.6 &10.8 &16.5 &21.0 &37.7 &2.4 & 8 \\
629: 030329 & 0.1685 & 0.48 &0.083 &1200 &1400 &1770 &1860 &2530 &3300 &\nodata &\nodata &0.7 & 9 \\
630: 030429 & 2.658 & 0.63 &0.204 &\nodata &\nodata &11.2 &16.3 &21.1 &34.3 &\nodata &57.8 &0.548& 10 \\
631: \tableline
632: \end{tabular}
633: \tablecomments{Data are all taken from the cited references.
634: The Galactic visual extinction $A_V^{\rm MW}$ is derived from
635: the reddening maps of Schlegel et al.\ (1998).
636: The UBVRIJHK fluxes are measured in the observer frame;
637: $z$ is the redshift of the burst; $t$ is the time when the data was
638: taken since the burst; and $\beta_0$ is the intrinsic spectral index
639: derived from the standard afterglow model (see \S2).} \tablerefs{(1)
640: Frail et al.\ 2003; (2) Jensen et al.\ 2001; (3) Jakobsson et al.\
641: 2003; (4) Stratta et al.\ 2005; (5) Covino et al.\ 2003; (6) Holland
642: et al.\ 2003; (7) Klose et al.\ 2004; (8) Vreeswijk et al.\ 2004;
643: (9) Bloom et al.\ 2004; (10) Jakobsson et al.\ 2004.} }
644: \end{table}
645:
646: \clearpage
647:
648: \begin{table}
649: \caption[]{Extinction in the rest frame of each burst
650: calculated from Table 1. $A_V$ is obtained by extrapolating
651: an analytical fit of $A_\lambda-A_{\left[V(1+z)\right]}$ to
652: $\lambda \rightarrow \infty$ (see \S2).\label{tab:tbl2}}
653: {\footnotesize
654: \begin{tabular}{ccccccccccccc}
655: \tableline\tableline
656: GRB & $z$ & $\beta_0$ & $A_V$ &\multicolumn{8}{c}{$A_\lambda-A_{[V(1+z)]}$}\\
657: \cline{5-12}
658: & & & & $U$ & $B$ & $V$ & $R$ & $I$ &$J$ &$H$ & $K$\\
659: \cline{4-11} \tableline
660: 980703 & 0.966 & 1.07 &0.48 &\nodata &0.29 &0.29 &0.24 &0 &0 &-0.41 &-0.26 \\
661: 000301C & 2.0404 & 0.56 &0.35 &0.51 &0.43 &0.34 &0.26 &0.21 &0.1 &0.01 &-0.13 \\
662: 011211 &2.14 &0.63 &0.07 &0.18 &0.12 &0.13 &0.07 &0.09 &-0.002& \nodata &0.002 \\
663: 020405 &0.691 & 0.5 &0.57 &0.66 &0.47 &0.30 &0.28 &0.04 &-0.39 &-0.45 &-0.51 \\
664: 020813 & 1.255 & 0.52 &0.43 &0.65 &0.52 &0.48 &0.39 &0.31 &-0.01 &-0.26 &-0.09 \\
665: 021004 &2.33 &0.45 &0.51 &1.42 &0.99 &0.65 &0.55 &0.43 &\nodata &0.06 &\nodata \\
666: 030226 &1.986 &0.33 &0.40 &\nodata &0.45 &0.36 &0.28 &0.22 &0.09 &-0.002&-0.20 \\
667: 030323 & 3.3718 &0.28 &0.92 &\nodata &2.64 &1.71 &1.29 &0.99 &0.84 &0.64 &0.15 \\
668: 030329 &0.1685 &0.5 &0.36 &0.14 &0.08 &-0.05 &0.03 &-0.16 &-0.26 &\nodata &\nodata \\
669: 030429 &2.658 &0.63 &0.41 &\nodata &\nodata &0.74 &0.52 &0.42 &0.14 &\nodata &-0.03 \\
670: \tableline
671: \end{tabular}
672: }
673: \end{table}
674:
675: \clearpage
676:
677: \begin{table}
678: \caption[]{Parameters for fitting the extinction curves of
679: GRB host galaxies with the Fitzpatrick \& Massa
680: (1990) formulae. Note that $x_0$ is fixed at
681: $4.6\mum^{-1}$ and is not a free parameter
682: (see Eq.[1]).\label{tab:tbl3}}
683: \begin{tabular}{ccccccc}
684: \tableline\tableline
685: & $c_1$ & $c_2$ & $c_3$ & $c_4$ & $\gamma$ ($\mum^{-1}$) & $k$ \\
686: \cline{1-7}
687: Type-I & -0.342 & 0.181 & 331.86 &0.040 &16.040 &0.340 \\
688: Type-II & -0.343 & 0.263 & 268.34 &0.061 &14.431 &0.358 \\
689: \tableline
690: \end{tabular}
691: \end{table}
692:
693: \clearpage
694:
695: \begin{table}
696: \caption[]{Parameters for modeling the 2-type extinction curves
697: of GRB host galaxies (as well as that of MW, SMC and LMC)
698: with the silicate-graphite dust model.\label{tab:tbl4}}
699: \begin{tabular}{ccccc}
700: \tableline\tableline
701: Galaxy & $\eta$ & $a_c$ ($\mu$m) & $f_{\rm gra}$
702: & $\chi^2/{\rm dof}$\tablenotemark{a}\\
703: \cline{1-5}
704: MW & 3.14 & 0.17 & 0.70 & 0.30 \\
705: LMC & 3.16 & 0.20 & 0.29 & 0.04 \\
706: SMC & 3.33 & 0.26 & 0.02 & 0.06 \\
707: Type-I & 2.61 & 0.21 & 0 & 0.63 \\
708: Type-II & 3.09 & 0.29 & 0.22 & 1.97 \\
709: \tableline
710: \end{tabular}
711: \tablenotetext{a}{\footnotesize $\chi^2 \equiv \sum_{\rm all\,\lambda}\sum_{\rm all\,GRBs}\left[ \left(A_\lambda/A_V\right)_{\rm mod} - \left(A_\lambda/A_V\right)_{\rm obs}\right]^2/\sigma^2$ is obtained by summing up
712: all wavebands and all GRBs, where $\sigma$ is the uncertainty
713: for a given GRB at a given band.}
714: \end{table}
715: \end{document}
716: