High density tracers of molecular gas in early-

type galaxiesAlison Crocker (UMass Amherst)

and Melanie Krips, Martin Bureau, Katey Alatalo, Leo Blitz,

Timothy Davis, Lisa Young and the Atlas3D team, Estelle Bayet

Stereotypical early-type galaxy (E or S0) = “red and dead”: no star formation, no cold gas

But now lots of evidence that this picture does not hold:

Introduction

HI: ½ of all field ETG

CO: ¼ of ETG Optically blue: 6% of ETG

UV blue: 30% of ETG

(Grossi+ 09, Oosterloo+ 10)‏

(Combes+ 07, Young+ 11)‏

(Yi+ 05)‏ (Schawinski+ 09)‏

Stereotypical early-type galaxy (E or S0) = “red and dead”: no star formation, no cold gas

But now lots of evidence that this picture does not hold:

Atlas3D project- a multi-wavelength volume-selected survey of 260 early-type galaxies (IFU spectroscopy, optical photometry, CO (single-dish + interferometry), HI, FIR)

Introduction

HI: ½ of all field ETG

CO: ¼ of ETG Optically blue: 6% of ETG

UV blue: 30% of ETG

(Grossi+ 09, Oosterloo+ 10)‏

(Combes+ 07, Young+ 11)‏

(Yi+ 05)‏ (Schawinski+ 09)‏

As part of Atlas3D, a 12CO(1-0) and (2-1) survey was conducted at the IRAM 30m

detected 56/260 (=22%), no trend with mass or environment (see Davis poster, Young+ 11)

no slow rotators with CO (the new ‘true’ ellipticals)

Introduction

As part of Atlas3D, a 12CO(1-0) and (2-1) survey was conducted at the IRAM 30m

detected 56/260 (=22%), no trend with mass or environment (see Davis poster, Young+ 11)

no slow rotators with CO (the new ‘true’ ellipticals)

NEXT STEP: What is the state of the molecular gas in early-type galaxies? Unique, relevant properties:

1. Very central CO distributions → shear important?

2. Significant hot gas → strong confinement by hydrostatic pressure?

3. Gas not being replenished (some very low H2-to-HI ratios)

Introduction

Observed ~1/3 brightest (=18) 12CO detections in:

13CO(1-0), 13CO(2-1), HCN(1-0), HCO+(1-0)

at IRAM 30m

Observations and Sample

Sample - H2 masses: 108 - 109 M⊙ - Avg. H2 surface densities: ~10-100 M⊙ pc-2

- Stellar masses: ~1010-1011 M⊙ - Mix of cluster, field, group

Hope that:

12CO/13CO indicates average optical depth of CO

HCN/13CO indicates dense gas fraction

Of course, complications-Abundance issues: 12C/13C decreases with chemical enrichment, HCN influenced by AGN(?)

Fractionation towards 13CO at cold gas temps (lower 12CO/13CO)

Selective photodissociation of 13CO in PDRs (higher 12CO/13CO)

Plan

NGC3032

VLSR (km/s)

T mb (

mK

)

-20

0

20

40

6080 12CO(1-0)

-20

0

20

40

6080 12CO(2-1)

-4-202468

10 13CO(1-0)

-5

0

5

10 13CO(2-1)

-2-1

0

1

23 HCN(1-0)

500 1000 1500 2000-6-4

-2

0

24 HCO+(1-0)

NGC2764

VLSR (km/s)

-20

0

20

40

60 12CO(1-0)

-40-20

020406080

100 12CO(2-1)

-4-2

0

2

46 13CO(1-0)

-5

0

5

10 13CO(2-1)

-2

-1

0

1

2 HCN(1-0)

2000 2500 3000 3500-3-2-10123 HCO+(1-0)

UGC09519

VLSR (km/s)

-20

0

20

40

6080 12CO(1-0)

-20

0

20

40

6080 12CO(2-1)

-4

-2

0

2

4 13CO(1-0)

-10

-5

0

5

10 13CO(2-1)

-3-2

-1

0

12 HCN(1-0)

1000 1500 2000 2500-2-1

0

1

23 HCO+(1-0)

NGC5866

VLSR (km/s)

-10

0

10

20

3040 12CO(1-0)

-10

0

10

20

3040 12CO(2-1)

-4-202468 13CO(1-0)

-5

0

5

10

15 13CO(2-1)

-3-2-10123 HCN(1-0)

0 500 1000 1500-2-1

0

1

23 HCO+(1-0)

X

X

NGC3032

VLSR (km/s)

T mb (

mK

)

-20

0

20

40

6080 12CO(1-0)

-20

0

20

40

6080 12CO(2-1)

-4-202468

10 13CO(1-0)

-5

0

5

10 13CO(2-1)

-2-1

0

1

23 HCN(1-0)

500 1000 1500 2000-6-4

-2

0

24 HCO+(1-0)

NGC2764

VLSR (km/s)

-20

0

20

40

60 12CO(1-0)

-40-20

020406080

100 12CO(2-1)

-4-2

0

2

46 13CO(1-0)

-5

0

5

10 13CO(2-1)

-2

-1

0

1

2 HCN(1-0)

2000 2500 3000 3500-3-2-10123 HCO+(1-0)

UGC09519

VLSR (km/s)

-20

0

20

40

6080 12CO(1-0)

-20

0

20

40

6080 12CO(2-1)

-4

-2

0

2

4 13CO(1-0)

-10

-5

0

5

10 13CO(2-1)

-3-2

-1

0

12 HCN(1-0)

1000 1500 2000 2500-2-1

0

1

23 HCO+(1-0)

NGC5866

VLSR (km/s)

-10

0

10

20

3040 12CO(1-0)

-10

0

10

20

3040 12CO(2-1)

-4-202468 13CO(1-0)

-5

0

5

10

15 13CO(2-1)

-3-2-10123 HCN(1-0)

0 500 1000 1500-2-1

0

1

23 HCO+(1-0)

X

X

18/18 detected in 13CO(1-0) and (2-1)12/18 detected in HCN(1-0)8/18 detected in HCO+(1-0)

Order-of-magnitude range (3-30)

(1-0) and (2-1) ratios well correlated

no huge differences in excitation or beam filling

12CO/13COOptically thinOptically thick

1012CO/13CO(1-0)

10

12CO/13CO(2-1)

NGC3665

NGC1266

NGC4459

NGC4526

E and S0s mostly seen in same range as spirals, Seyferts and starbursts

But 3 galaxies have extremely low ratios, similar to what is observed for MW GMCs! (No diffuse molecular gas?)

12CO/13CO

1012CO/13CO(1-0)

10

12CO/13CO(2-1)

1012CO/13CO(1-0)

10

12CO/13CO(2-1)

NGC3665

NGC1266

NGC4459

NGC4526

E/S0SpiralStarburstSeyfert

Optically thinOptically thick

Compilation of Seyferts and starbursts from Krips et al. 2010; spiral range from Paglione

et al. 2001

Galaxies with the lowest 12CO/13CO ratios are in the most luminous (massive) galaxies

Some relation with dust (double branch: starburst and not)

No clear dependence on environment or H2/HI ratio

12CO/13CO

-25 -24 -23 -22MK

1

10

100

12CO

/13CO

(1-0

)

0.2 0.4 0.6 0.8 1.0f60/f100

-3 -2 -1 0 110th nearest gal. (Mpc-3)

-1 0 1 2 3log(H2/HI)

Galaxies with the lowest 12CO/13CO ratios are in the most luminous (massive) galaxies

Some relation with dust (double branch: starburst and not)

No clear dependence on environment or H2/HI ratio

12CO/13CO

-25 -24 -23 -22MK

1

10

100

12CO

/13CO

(1-0

)

0.2 0.4 0.6 0.8 1.0f60/f100

-3 -2 -1 0 110th nearest gal. (Mpc-3)

-1 0 1 2 3log(H2/HI)

Why? Some options:

low in coldest-dust galaxies -> fractionation

low in most massive/highest metallicity -> abundance

low in most massive/lowest turbulence gas -> optical depth

12CO/13CO

-25 -24 -23 -22MK

1

10

100

12CO

/13CO

(1-0

)

0.2 0.4 0.6 0.8 1.0f60/f100

-3 -2 -1 0 110th nearest gal. (Mpc-3)

-1 0 1 2 3log(H2/HI)

Comparison to spirals (Paglione+01)

Overlap with dust temp, not with mass (if fractionation partly to blame, also working similarly in spirals)

12CO/13CO

-25 -24 -23 -22MK

1

10

100

12CO

/13CO

(1-0

)

0.2 0.4 0.6 0.8 1.0f60/f100

-3 -2 -1 0 110th nearest gal. (Mpc-3)

-1 0 1 2 3log(H2/HI)

Consider HCN/13CO as tracing the fraction of dense molecular gas

Smallish range 0.1-0.5

Also the unusual NGC1266 (molecular outflow galaxy, see Alatalo poster!)

HCN/COHigh dense gas fraction

Low dense gas fraction

0.1 1.0HCN/13CO(1-0)

0.01

0.10

1.00

HCN/12CO(1-0)

NGC3607NGC1266

NGC4459NGC4526

E and S0s are to the low side- lower than starbursts, Seyferts, possibly spirals

Thus inefficient at making densest gas.

3 optically thick galaxies again ‘short’ on 12CO

HCN/COHigh dense gas fraction

Compilation of Seyferts and starbursts from Krips et al. 2010

0.1 1.0HCN/13CO(1-0)

0.01

0.10

1.00

HCN/12CO(1-0)

0.1 1.0HCN/13CO(1-0)

0.01

0.10

1.00

HCN/12CO(1-0)

NGC3607NGC1266

NGC4459NGC4526

E/S0SpiralStarburstSeyfert

Low dense gas fraction

Again, massive galaxies low in 12CO- not just a 12C vs 13C effect!

Most likely culprit: high optical depth in 12CO on a global scale for these galaxies.

Same basic pattern with dust temp (except 1266)

12CO/HCN

-25 -24 -23 -22MK

10

100

12CO

/HCN

(1-0

)

0.2 0.4 0.6 0.8 1.0f60/f100

-3 -2 -1 0 110th nearest gal. (Mpc-3)

-1 0 1 2 3log(H2/HI)

Again, massive galaxies low in 12CO- not just a 12C vs 13C effect!

Most likely culprit: high optical depth in 12CO on a global scale

Is correlation with mass or dust temp more fundamental, or all related to lower turbulence/increased self-absorption?

12CO/HCN

-25 -24 -23 -22MK

10

100

12CO

/HCN

(1-0

)

0.2 0.4 0.6 0.8 1.0f60/f100

-3 -2 -1 0 110th nearest gal. (Mpc-3)

-1 0 1 2 3log(H2/HI)

Add ‘normal’ spirals from Gao & Solomon 04 (< 1010 Lsun in FIR)

HCN/12CO

-25 -24 -23 -22MK

10

100

12CO

/HCN

(1-0

)

0.2 0.4 0.6 0.8 1.0f60/f100

-3 -2 -1 0 110th nearest gal. (Mpc-3)

-1 0 1 2 3log(H2/HI)

No clear correlation with luminosity, dust temp, environment or H2/HI ratio for HCN/13CO

No uniform spiral comparison sample

HCN/13CO

-25 -24 -23 -22MK

1

10

13CO

/HCN

(1-0

)

0.2 0.4 0.6 0.8 1.0f60/f100

-3 -2 -1 0 110th nearest gal. (Mpc-3)

-1 0 1 2 3log(H2/HI)

more dense

less dense

HCN/HCO+

0.1 1.0HCN/13CO(1-0)

1

HCN/HCO

+ (1-0)

0.1 1.0HCN/13CO(1-0)

1

HCN/HCO

+ (1-0)

E/S0StarburstSeyfert

Wider spread in HCN/HCO+ than seen in local GMCs (pink range), but big error bars

Effect of AGN for some, but not all have an AGN...

Some other chemical effect?

Most early-type galaxies broadly resemble spirals in terms of 12CO/13CO, HCN/13CO and HCN/12CO ratios

More luminous (massive) early-types have more optically thick 12CO (possible effect of lower turbulence)

Dust temp also correlates with diffuse molecular content (cold dust, less diffuse), but double-branched

Some interesting HCN/HCO+ ratios, but need to be better understood

Conclusions

LVG modeling by Bayet (see poster for first results!)

Herschel line studies of these galaxies (SPIRE FTS + PACS spectroscopy)

ALMA will enable resolved studies of these high-density tracers in ETG

ALMA’s increased sensitivity will also allow us to probe below these brightest (and most peculiar) ETG to ‘normal’ star forming ETG

Of course, need sample readjustment to South...

Other work and ALMA

HCO+

107 108 109 1010

L12CO(1-0) (K km s-1 pc-2)

109

1010

1011

1012

L FIR

(LO •)

Slope = 1.3

106 107 108 109 1010

LHCN(1-0) (K km s-1 pc-2)

Atlas3D E/S0sGao & Solomon 2004

Slope = 1.0