Monique  C.  Aller  (Georgia  Southern  University)    

Varsha  P.  Kulkarni  (University  of  South  Carolina)  Eli  Dwek  (NASA-­‐GSFC)  

 Donald  G.  York  &  Daniel  E.  Welty  (University  of  Chicago)  Giovanni  Vladilo  (Osservatorio  Astronomico  di  Trieste)  

¡  Introduction  –  Quasar  Absorption  Systems  (QASs)  §  How  we  can  use  QASs  to  study  dust  in  distant  galaxies  §  Evidence  for  dust  in  QASs  

¡     Silicate  dust  in  QASs  §  Dedicated  Spitzer  IRS  mini-­‐survey  to  study  silicate  dust  in  gas  and  dust-­‐rich  QASs  è  results,  trends  

§  NASA-­‐ADAP  ongoing  archival  study  to  silicate  dust  in  moderately  gas-­‐rich  QASs  with  archival  Spitzer  IRS  quasar  spectra  ècurrent  results,  open  questions  

¡     Summary  &  Future  Work  

Peeples  2015    

Impact  param

eter  

Cartoon  of  QAS  

Sightlines  to  distant  quasars  pass  through  foreground  (still  distant)  galaxies  

• Galaxies  sampled  by  gas  cross-­‐section,  independent  of  galaxy  brightness  • Primary  neutral  gas  reservoir  for  star-­‐formation  (e.g.,  Storrie-­‐Lombardi  &  Wolfe  2000):  

• Damped  Lyman-­‐α  absorbers  (DLAs):  NHI  ≥  2×1020cm-­‐2  

• sub-­‐DLAs:  1019≤NHI<2×1020cm-­‐2  

Credit:  A  Pontzen  

UV/Optical  Spectra  (gas  properties):  abundances;  kinematics;  temperature;  density;  ionization  parameter  

From  Webb;  Pettini  2003  

Si  IV    (1393  Å)  

C  IV    (1548  Å)  

Lyman  limit  

Cartoon  of  QAS  

Infrared  Spectra  (silicate  dust):  grain  property  constraints  from  10  and  18  μm  absorption  features  

York  et  al.  2006  

809  SDSS  absorbers  1<z<2    

SMC-­‐like  reddening  for  absorbers  

Statistical  evidence  of  reddening  for  quasars  with  foreground  

absorption  systems  è  dust  in  QASs    

– 23 –

Zn Si Mn Fe Cr Ti

[X/H

] = lo

g (X

/H) -

log

(X/H

) sun

Solar&Level&(0%&dust)&

-2.5 -2 -1.5 -1 -0.5 0 0.5Metallicity [X/H]

-1.5

-1

-0.5

0

Dep

letio

n [F

e/X

]

Fig. 1.— Left panel: (a) Abundances of a few key elements in DLAs and sub-DLAs, based

on data from Meiring et al. (2009) and references therein. Also shown for comparison are

the abundance patterns observed for the Galactic halo ISM, warm Galactic ISM, and the

SMC ISM. The depletion patterns are similar for DLAs and sub-DLAs, except the pattern

for sub-DLAs is o↵set to higher abundances. Both the DLA and sub-DLA depletion patterns

appear to be roughly similar to the Galactic halo ISM pattern. Right panel: (b) Depletion

of Fe relative to Zn or S (if Zn is not available) vs. the metallicity based on Zn or S, for the

sample of quasar DLAs analyzed in Kulkarni et al. (2015). A trend of increasing depletion

level with increasing metallicity is seen.

X  =  Zn  or  S  (if  Zn  unavailable)  

Kulkarni  et  al.  2016  

Dust  depletion  appears  anti-­‐correlated  with  metallicity    (e.g.,  Ledoux  et  al.  2003;  Meiring  et  al.  2006,  2009)    

Noterdaeme  et  al.  2009  

Dust  extinction  is  present  based  on  2175  Å  bump  (carbonaceous)  

zabs=1.64  like  LMC2    

estimated by varying E(B! V ) and comparing, by eye, thedepth of the resulting 2175 8 feature to the observed data. Therange of values found is from E(B! V ) ¼ 0:12 to 0.19, andthis range is used later to estimate part of the systematic errorsin the parameters to fit 1 (Table 1), which is our best estimate

for the dust parameters at za ¼ 0:524 along the AO 0235+164line of sight.Galactic-type dust extinction at za ¼ 0:524 produces the

best fit to the AO 0235+164 spectrum when comparing Ga-lactic-, SMC-, and LMC-type dust. The observed spectrum

Fig. 5.—HST STIS spectra of AO 0235+164 from Fig. 1 and Galactic, LMC, and SMC dust model !2 fits. The data are binned 2:1. For clarity, the continua of thefitting functions are shown displaced above the data. Superposed on the data is the Galactic model.

TABLE 1

AO 0235+164 Spectral Fits with Dust Extinction

Fit Dust Type Referencea E(B !V )b RVb "c, b

k1d

(8)k2

d

(8) !2#

1.............. Galactic 1 0.227# 0.003 2.51# 0.03 !1.75# 0.01 2500 8700 2.67

2.............. LMC 2 0.281# 0.004 3.16 !1.40# 0.02 2500 8700 5.31

3.............. SMC 2 0.161# 0.006 2.93 !1.63# 0.03 2500 8700 14.2

4.............. LMC 2 0.396# 0.005 3.16 !0.72# 0.03 2800 6000 2.50

5.............. Mod. Gal.e 3 0.211# 0.002 3.08 !1.76# 0.01 1690 8700 3.08

6.............. Gal. + LMCf 1+2 0.267# 0.003 2.95# 0.02 !1.60# 0.01 1690 8700 3.72

7.............. Galactic 1 0.223# 0.003 2.43# 0.03 !1.75# 0.01 2549 8700 2.44

7b............ . . . . . . . . . . . . +0.28# 0.06 1690 2549 . . .

a Reference for extinction formulae: (1) CCM89; (2) Pei (1992); (3) Pei (1992) modified to allow a variable far-UVextinction component. RV is a fixed value in the models using the Pei (1992) formulae for extinction and a variable inthe models using the CCM89 formulae.

b Only the formal error; systematic errors probably dominate and are much larger.c Power-law spectral index for the continuum modeled as f# / # " . For model 7 a broken power law was used, and

the continuum parameters for the blue part are given in line 7b.d Beginning and end wavelength of the spectral region modeled as a power law and included in the fit.e A modified Galactic extinction with a reduced far-UV component (see text).f A model with both Galactic- and LMC-type absorbers at za ¼ 0:524. The best fit found required no LMC-type

absorption [E(B! V ) $ 0:005].

JUNKKARINEN ET AL.664 Vol. 614

zabs=0.524  Galactic    

Junkkarinen  et  al.  2004  

15σ    (in  EW)  

detection  of  silicate  dust  in  z=0.524    

DLA  using  Spitzer  IRS  Feature  best-­‐matched  by  profiles  for  diffuse  Galactic  

interstellar  clouds  or  laboratory  amorphous  olivine  –  with  a  shallow  peak  optical  depth  (~0.08)  

Kulkarni  et  al.  2007  

¡  Select 12 additional QASs abundant in gas (and dust): §  Expected to be DLAs or sub-DLAs; i.e. most NH1>1020 cm-2

§  Signatures of dust à 2175 Å bump detected; reddening; large extinctions (AV >0.4); and/or gas-phase depletions

§  Signatures of molecular absorption in a few QASs (PKS 1830-211, TXS 0218+357)

¡  Spitzer  IRS    spectra  (PI  proposals:  V.P.  Kulkarni)  §  Data  in  low-­‐resolution  mode  (SL,  LL)  §  Cover  10  µm feature  every  QAS;  (part  of)  18  µm feature  in  6  QASs  

¡  QASs redshifts sample 0.156≤zabs≤1.388; quasars 0.87≤zem≤2.51  

10  µm  silicate  dust  

absorption  detected  at  >4σ  in  each  

system  

Kulkarni  et  al.  2011  

Grav.  Lensed  Blazar  (z=2.507)    

Absorber  (z=0.886):  Ø  Face-­‐on  (Sb/Sc)  spiral  galaxy  Ø Molecule-­‐rich:  CO,  HCO+,  HCN,  HS,  H2O,  NH3  

Ø  Neutral  gas  rich:  21-­‐cm  (HI)  absorption   10  µm absorption  

Aller  et  al.  2012  

18  µm  absorption  

HST  F814W  image,  Winn  et  al.  2002  

galaxy  lensed

 quasar  

12.7σ  (in  EW)  detection  of  10  µm  feature  

(=7.78  kpc  QAS)  

Examined  fits  with  >100  optical  depth  templates  (observed  &  lab  sources)    è  laboratory  crystalline  olivine  (Mg2xFe2-­‐2xSiO4)  silicates  è  unusually  high  crystallinity  (Galactic  ISM  =  amorphous)  

PKS  1830-­‐211  

Amorphous  olivines  provide  poor  fit  

(x=0.94)   (x=1.0)  

(x=0.0)  (x=0.55)  (Mg,Fe)3Si2O5(OH)4   SiO2  

Aller  et  al.  2012  

Aller  et  al.  2014  

Grav.  Lensed  Blazar  (z=0.944)  Absorber  (z=0.685):  Ø  Face-­‐on  spiral  galaxy  Ø Molecule-­‐rich:  CO,  HCO+,  HCN,  H2O,  NH3,  LiH  

Ø Neutral  gas  rich:  21-­‐cm  (HI)  absorption  

HST  imaging,  Jackson  et  al.  2000  

10 µm 18 µm

Sep:  335mas~2.4  kpc  in  QAS  

Detections  of  both  the  10µm and  the  18µm  silicate  absorption  features:  ¡  Significant  detections  -­‐10µm (10.7-σ in EW) and  18µm (>3σ) abs. ¡  Highest  peak  optical  depth  (τ10=0.49±0.02) for  any  QAS  so  far  

AMORPHOUS  OLIVINE  

Aller  et  al.  2014  

Like  PKS  1830-­‐211  is  gravitationally-­‐lensed,  molecule-­‐rich  BUT  unlike  PKS  1830-­‐211  BOTH  the  10  and  18  µm  absorption  matched  by  amorphous  olivine  profile  

Comparing  10  micron  absorption  feature  in  different  QASs:  

• Variations  in  depth  of  feature  • Variations  in  breadth  of  feature  • Variations  in  peak  wavelength  • Variations  in  substructure  

è  may  suggest  variations  in  silicate  dust  grain  properties  between  different  QASs    è  variations  do  not  clearly  correlate  with  other  QAS  parameters  like  presence  of  molecules,  absorber  redshift  

Aller  et  al.  2014b  

Milky  Way  diffuse  clouds

 extrapolation  

• Grain  differences?    e.g.  larger  grains-­‐low  UV  extinction    • Sampling  different  dust  grain  population?  e.g.  face-­‐on  vs.  through  MW  disk    • Different  stellar  populations?    e.g.  more  O-­‐rich  

Slope  of  relation  is  3-­‐6x  higher  for  QASs  than  for  Milky  Way  diffuse  clouds  

Aller  et  al.  2014  

fcov=1  assumed  in  all  QASs  

Kulkarni  et  al.  2011  

Hint  of  an  anti-­‐correlation  between  carbonaceous  and  

silicate  dust  absorption  strengths  

   à  is  dust  

predominately  carbonaceous  or  silicate  along  a  

sightline?  

Aller  et  al.  2014  

Suggestion  of  trend  between  Mg  II  EW  and  silicate  10  µm  peak  optical  depth:  

 

Are  silicate-­‐rich  QASs  more  massive?  

Mg  II  saturated  in  most  systems,  and  is  proxy  for  velocity  spread  (QAS  mass;  outflows)  

TXS  0218+357  

 

DUST  MEASUREMENTS  IN  QASS:  ¡  Silicate  dust  10  µm  feature  in  ≥50  QASs  toward  quasars  with  Spitzer  IRS  data  ¡  Signatures  of  silicate  crystallinity  (substructure,  longer-­‐λ  resonance  features)?  ¡  Determine  shapes  of  extinction  curves  for  ≥70  individual  QASs  ¡  Examine  relative  abundances  of  carbon:silicate  dust  (~35  QASs)  

GAS  MEASUREMENTS  IN  QASS  WITH  DUST  INFORMATION:  

¡  Ascertain  gas  metallicity  and  depletions  (e.g.  Fe,  Cr,  Si)  ¡  Estimate  gas  kinematics  (e.g.  velocity  structure,  widths)  

CONNECTIONS  BETWEEN  DUST/GAS  AND  MODELS:  ¡  Examine  interrelation  between  gas  and  dust  properties  in  connection  to  dust/  chemical  evolution  models    V.  P.  Kulkarni  (PI,  U.  South  Carolina);  M.C.  Aller  (co-­‐I/Science-­‐PI;  Ga.  Southern  U.);  E.  Dwek  (co-­‐I;  NASA-­‐GSFC)  

¡  Quasars  with  archival  Spitzer  IRS  spectra  §  Most  in  low  resolution  (SL,LL)  stare  mode  §  Several  mapped,  higher  resolution  spectra  

¡  Cataloged  foreground  absorbers  with  moderate  gas  §  Log  NHI≥18.0  (most  not  DLAs  or  sub-­‐DLAs,  median=18.23)  §  EWMg  II≥0.5  Å  (median  ~0.9)  or  EWNa  I≥1.0  Å    §  Absorbers  10  μm  or  18  μm  silicate  feature  would  fall  within  observed  spectrum  

¡  QASs  are  purely  gas-­‐selected  (no  prior  requirements  for  evidence  of  QAS  dust  or  molecules)  

èfinal  sample  of  37  quasars  (some  with  multiple  absorbers)  §  median  zquasar=1.0  (0.3<zquasar<4.4)    §  median  zQAS=  0.7  (0.1<zQAS<2.8)  

SBS#09009+532#Spectrum#3C#175#Spectrum#

3C  175  •  z=0.77  Seyfert  (1.2)    •  Optically  variable  

Absorber  (zQAS=0.46)  •  NH  I<18.30  Rao  et  al.  2011    

•  EWMg  II=0.62  Å  Rao  et  al.2011  

3C  196  •  z=0.87  QSO  •  Optically  variable  

Absorber  (zQAS=0.44)  •  NH  I=20.8  (DLA)  Boisse  et  al.  1998  •  ~  0.8-­‐2.0  x  solar  metallicity  •  SFR  ~5  M⊙ yr-1 Gharanfoli et al. 2007

•  Barred  spiral    

SBS  0909+532  •  z=1.38  AGN  •  Grav.  Lensed  

Absorber  (zQAS=0.83)  •  NH  I≥21.14  (DLA)  •  EWMg  II=2.o;  0.7  Å  •  2175  Å  bump  Motta  et  al.  2012    •  Early  type  galaxy  

3C#196#Spectrum# SBS  0909+532  Spectrum  3C  196  Spectrum  3C  175  Spectrum  

Aller  et  al.  in  prep  

3C#175#QAS#Amorph.#Olvine# 3C#175#QAS#Amorph.#Pyroxene# SBS0909+532#QAS#Amorph.#Olvine#

Some  uncertainty  on  the  intrinsic  shape  of  the  quasar  continuum  (limited  continuum  bracketing  absorption  features)  ècurrently  

exploring  what  range  of  grain  properties  would  be  consistent  with  plausible  quasar  continuum  normalizations      

τ10=0.32    τ10=0.28    τ10=0.19    

¡  Silicate  grain  variations?  

¡  Silicate  absorption  (QAS)  blended  with  silicate  emission  (quasar)  in  some  systems?  

¡  Differences  in  sightline  to  quasar?  

3C#175#QAS#Amorph.#Olvine#

What  is  the  origin  of  these  poor(er)  fits?  

¡  Silicate  grain  variations?  

¡  Silicate  absorption  (QAS)  blended  with  silicate  emission  (quasar)  in  some  systems?  

¡  Differences  in  sightlines  èPKS  1830-­‐211;  AO  0235+164;  TXS  0218+357  are  blazars  è  Systems  with  poor(er)  fits  are  not  blazars;      sightline  intersecting  dust  torus?  

Credit:  Pierre  Auger  Observatory  

¡  Silicate  dust  in  absorption  in  QASs    §  Evidence  of  10  µm  (and  18  µm)  absorption  seen  in  QASs  at  z<1.4  with  

strong  low-­‐ion  metal  absorption  lines  and/or  signatures  of  dust  §  Variation  in  shape,  breadth  of  absorption  feature      

¡  Trends  of  τ10 with  other  dust  and  gas  properties  of  QAS  §  Correlation  with  E(B-­‐V)  –  but  steeper  slope  than  in  MW  clouds  §  Correlation  with  Mg  II  EW  –  silicate  rich  are  more  massive?  §  Suggestion  of  anti-­‐correlation  with  carbonaceous  dust  abs.  strength  

¡  Some  systems  show  evidence  of  possible  silicate  emission  in  quasar  (as  well)  §  Currently  exploring  more  sophisticated  quasar  continuum  

normalizations  accounting  for  this  

¡  Final  analysis  with  expanded  sample  will  better  constrain  silicate  dust  trends,  and  explore  prevalence  of  silicate  dust  in  distant  galaxies