as seen with MIDI - Max Planck Society · 2014. 5. 14. · IntroThe MIDI-Herbig projectFirst...

Intro The MIDI-Herbig project First results Conclusions

The structure of Herbig-star disksas seen with MIDI

J. Menu, C. Waelkens

Institute for Astronomy, KU Leuven [B]

R. van Boekel, Th. Henning

Max Planck Institute for Astronomy, Heidelberg [D]

1 / 23


Intro: protoplanetary disks & MIDI

Disk structure

E. Di Folco et al.: The flared inner disk of the Herbig Ae star AB Aurigae revealed by VLTI/MIDI in the N-band 1073

34 m

8 9 10 11 12 13

0.0

0.2

0.4

0.646 m

8 9 10 11 12 13

0.0

0.2

0.4

0.670 m

8 9 10 11 12 13

0.0

0.2

0.4

0.6

84 m

8 9 10 11 12 13

0.0

0.2

0.4

0.6116 m

8 9 10 11 12 13

0.0

0.2

0.4

0.6 MIDI Visibilities

MC3D best fit for :

MIDI + ISO (major axis)

MIDI + ISO (minor axis)

MIDI (Vis + sed)

MIDI Vis only

λ [µ m]

Vis

ibili

ty

λ [µ m] λ [µ m]

λ [µ m]

Vis

ibili

ty

λ [µ m]

Fig. 7. MIDI observations compared to the best-fit models presented in Figs. 5 and 6. Squares (triangles) stand for the simulated visibilities alongthe major (resp. minor) axis of the modelled disk. VLTI observations are presumably aligned with the major axis. We present the best-fit modelsfor the interferometric constraint alone (dots) and the combination of MIDI visibilities with the MIDI spectrum (diamonds) or the ISO spectrum(squares and triangles). The visibility constraint has a greater weight than the spectrum in the fitting process (see text).

h100= 8 AU, β= 1.267

h100 +/− 2.0 AU

β +/− 0.025

0.1 1.0 10.0

0.01

0.1

Radial distance r [AU]

Vert

ical scale

heig

ht h [A

U]

Fig. 8. Variation in the scale height in the very inner disk for the best-fitmodel presented in Fig. 5 and influence of a small change of h100 and β.The inner rim is described by a simple vertical wall.

of β in the range 1.25–1.30, slightly higher than those inferredfrom the SED-fit alone. We show in the next section that thisestimation is quite robust, even when small variations of themodel input parameters are considered. Piétu et al. (2005) de-rived β = 1.4 from the 13CO maps, a value significantly higherin the outer regions of the disk (80–1300 AU). The associatedscale height at 100 AU is very close to our best-fit value withh100 = 8.5 AU3. If assuming a Keplerian velocity field in the in-ner regions of the disk (not completely guaranteed in the caseof AB Aur if we extrapolate to smaller radii the conclusions ofPiétu et al. 2005), the radial gradient of temperature in the diskphotosphere probed by our mid-IR observations is T ∝ r−q withq = 3 − 2β ≃ 0.5−0.52 for β ≃ 1.25−1.30.

3 Their definition of the scale height differs from ours by a factor√

2.

Similarly, we can compare our results for the radial surfacedensity power index with that estimated by large-scale observa-tions. The assumed α/β combination yields α ≃ 2.25−2.40 andthus for the surface density (Σ(r) ∝ r−p) p = α − β ≃ 1.0−1.1.This value is slightly lower than the one usually adopted forthe young solar system (p = 1.5) and significantly lower thanthe index derived from mm observations for the outer annulus(p ≃ 2.3−2.7). Discarding the previously adopted α = 3(β−0.5)relation to fix p = 2.5, we have checked that the fit wouldlead to a little increased flaring index (β = 1.3) and to a largerscale height (h100 ∼ 10.5 AU), still in the same order as that de-rived from mm observations. It should, however, be emphasisedthat the above-mentioned mm interferometric observations haveshown a dramatic change in the disk properties inside ∼100 AU,so that extrapolating these morphological parameters to the innerregions might be quite hazardous.

5.2. Influence of input parameters

In this section, we investigate the dependence of the best-fit val-ues β and h100 on the main model input parameters to highlightpossible hidden degeneracies in our analysis. Figures 9 to 11present the best-fit values for the various models, each point be-ing derived from a combined χ2-map, like those presented inFig. 5c. The associated χ2

redtotare close to each other and the as-

sociated models are almost equally probable.

1. The dust mass. Mdust was varied from 5 × 10−6 to 1 ×10−4 M⊙. The upper value is in line with the typical massesused in the litterature to fit the mid-IR spectrum (e.g.,Bouwman et al. 2000), while the lower value correspondsto our estimation of the dust amount, that would produce the

Flaring disk of AB Aur(di Folco+ 2009) Rounded rim of TW Hya

(Menu+ 2014)

2 / 23


Intro: protoplanetary disks & MIDI

Disk populationsCh. Leinert et al.: Sizes of disks around Herbig Ae/Be stars 547

Fig. 5. Correlation between the mid-infrared spectral slope (taken

from IRAS as −2.5log (Fν(12 µm)/Fν(25 µm)) and the half-light ra-

dius corresponding to the observed visibilities at 12.5 µm. The errors

reflect the uncertainty in IRAS fluxes and visibility measurements; the

resulting uncertainty in size may be optimistic. The largest sources

are those with the reddest mid-infrared spectral distribution, and these

were also classified as group I. For HD 142527 the arrow shows the

correction for contamination of the IRAS colour by an unrelated very

red nearby source (see text).

5.3.1. Description of the models

The DDN models refer to passive, centrally irradiated circum-

stellar disks with an inner hole (Dullemond et al. 2001). The

SED emerging from such a disk primarily has two main com-

ponents: (i) an optically thin emission from the surface layer,

responsible for the observed solid-state emission features and

part of the near-and mid-IR flux, and (ii) a component origi-

nating from the midplane of the disk which contributes to the

mid-IR flux and dominates the far-IR and sub-millimeter wave-

length regions. In the innermost regions of Herbig Ae/Be disks,

the temperatures become so high that dust grains evaporate. At

the interface between dusty and dust free regions, a puffed-up

inner rim forms. It intercepts up to 25% of the stellar radia-

tion and re-emits as a blackbody component of typically 1200–

1500 K. Just outside the inner rim, a shadow is cast over the

disk out to a radius of about 5–10 AU. The hot inner rim as

third component in the SED, effective in explaining the near-

infrared (2–7 µm) bumps observed in some Herbig Ae/Be stars,

is the essential new ingredient in the DDN models with respect

to earlier approaches (Kenyon & Hartmann 1987; Chiang &

Goldreich 1997). As in these older models, a flaring disk ge-

ometry in which the surface curves upwards as a function of

distance from the star is normally used.

With suitable parameters, these DDN models provide a rea-

sonable fit to the SEDs (star plus disk) of most Herbig Ae stars,

(Dominik et al. 2003), and they were constructed just for this

purpose. No spatial information has entered into the fitting

procedure. The fundamental stellar parameters are taken from

Meeus et al. (2001), while the disk is described by the param-

eters disk mass Mdisk, the slope of the surface density power

law p, the outer disk radius rout, the inclination i and in some

cases the height of the inner rim χ. Dust grains 0.1 µm in size

are assumed in a mixture of astronomical silicate (Draine &

Lee 1984) and carbon (Laor & Draine 1993). The contribution

by direct stellar flux is usually negligeable (Fig. 1). For most

programme stars, fits were already available (Dominik et al.

2003). For KK Oph, the fit to the spectral energy distribution

yielded the parameters summarised in Sect. 3.2.

5.3.2. Comparison of visibilities

The general observed trend of visibility with wavelength is

given by a pronounced decrease between 8 µm and 9 µm fol-

lowed by a plateau out to 13 µm. This trend appears even

clearer in the visibility predictions based on SED fitting DDN

models, which also are plotted in Fig. 4. The tentative explana-

tion for this behaviour is that the hot inner rim region of the cir-

cumstellar disks gives an overproportional contribution to the

shorter wavelength region, resulting in a smaller effective size

at these wavelengths than would be expected for the smooth

temperature distribution in the disk. We take the general agree-

ment between observations and predictions as qualitative con-

firmation of the physical picture underlying the DDN models.

However, quantitatively there remain significant differences be-

tween model predictions and observations to the extent that in

no case there is real good agreement between them. The ob-

served decrease in visibility is steeper (for HD 100546) or shal-

lower (for HD 142527, HD 144432, KK Oph) than predicted.

The level of the “plateau” longward of 9 µm is also different

in some cases. But these differences are usually less than 30%

of the observed visibility which will translate to smaller correc-

tions in size because of the non-linear transformation to visibil-

ities. The largest deviation is seen in HD 163296 with a factor

of 1.5–1.8 which translates into differences in the width of an

equivalent Gaussian distribution of 13% to 20%, which is not a

dramatic effect.

Certainly, one does not expect a perfect spatial distribution

of infrared emission predicted by models constructed solely to

fit the spectral energy distribution. But the discrepancies show

that these models cannot be the last word but will need modifi-

cations. It will be important to include spatial constraints such

as those given by the presented interferometric observations

into the next efforts of circumstellar disk modeling to reach

a deeper understanding of structure, physics and evolution of

such disks.

We do not try in this paper to produce such improved fits.

As the probably best example of a spatially and spectrally self-

consistent model for HL Tau (Men’shchikov et al. 1999) shows,

this tends to require an extensive effort, best to be done care-

fully on an object by object basis.

6. Summary and conclusion

Our mid-infrared interferometric study of a sample of seven

Herbig Ae/Be stars has shown that

– MIDI on the VLTI is well able to resolve the circumstellar

dust structures around Herbig Ae stars in the mid-infrared.

Sizes of “flaring” and “flat” disks (Leinert+ 2004)

3 / 23


Disk structure & MIDI

+well-defined disk structure for several targets

hints of different disk populations

–lack of statistics: case studies difficult to combine/compare

archive with lots of data, many objects untouched

4 / 23


The MIDI-Herbig project

〈〈 Collective data reduction and analysis of (almost) allMIDI data of intermediate-mass young stars 〉〉

Data: 239 nights of archival MIDI observations, 80 targets

First results:

1 size-luminosity relation

2 towards a family picture

5 / 23


First results: 1. Size-luminosity relation

Principle: complex disk → simple disk

R=Rsub

R=100 AU

temp.gradient

Fig left: Rice+ (2003); right: temperature-gradient disk

6 / 23



10-1 100 101 102 103 104 105 106 107

luminosity (L⊙)

10-2

10-1

100

101

102

half-light radius (AU)

Fig Mid-infrared (10-µm) size-luminosity relation

7 / 23



10-1 100 101 102 103 104 105 106 107

luminosity (L⊙)

10-2

10-1

100

101

102

half-lig

ht ra

diu

s (A

U)

T = 90

0 KT =

250 K

transition disk

debris disk

Fig Mid-infrared (10-µm) size-luminosity relation

8 / 23


Comparison with Monnier+ (2009)No. 1, 2009 MID-IR SIZES OF YSOs 503

100 102 104 106

Stellar Luminosity (solar)

1

10

100

Rin

g R

adiu

s (

AU

)

abaur fu

ori

(b)

hd142527

hd142666

hd144432(b

)

hd158643

hd41511

hd45677

hd50138

lkha101(b

)

lkha234(b

)

monr2

irs3(b

)

mw

c1080(b

)

mw

c275

mw

c297

mw

c300(b

)

mw

c342

mw

c349(b

)

mw

c614

mw

c863(b

)

rcra

rmon(b

)

ryta

uttaun(b

)

v1295aql

v1685cyg(b

)

v645cyg

v892ta

u(b

)

zcm

a(b

)

mw

c480

gw

ori

(b)

v376cas

v883ori

mw

c147(b

)

mw

c361(b

)

afg

l490

afg

l2136

afg

l2591

s140irs1

T=25

0K

T=90

0K

EmbeddedHerbig AeHerbig BeUncertainTTauriFU Ori

YSO mid-IR Size Luminosity Diagram

Figure 8. Mid-infrared size–luminosity diagram. Different plot symbols correspond to different YSO types: embedded YSOs (asterisk), Herbig Ae (X), Herbig Be(diamond), T Tauri (triangle), FU Orionis objects (square), and nonclassified emission line objects (plus). Systems in close binaries are labeled (b). The lines representdifferent temperatures for gray dust according to the definition in Monnier & Millan-Gabet (2002). Objects with FWHM sizes smaller than 35 mas are labeled asupper limits.

(A color version of this figure is available in the online journal.)

We also note that the embedded YSOs all cluster near the top-right of the diagram at high luminosities and low temperatures.This is expected since the high optical depths toward these ClassI objects mean their infalling envelopes are still optically thickand we do not expect to be able to see into the warmer innerregions surrounding these protostars. Note that these regions allshow complicated, often bipolar, structures (see Figure 6) andthat the sizes in this diagram refer only to the central core ofemission.

4.2. Size Versus IRAS Color

Leinert et al. (2004) found some evidence that the physicalemission scale for the mid-infrared emission (in Herbig Ae stars)was correlated with the IRAS 12–25 µm color, defined as −2.5log Fν(12 µm)/Fν(25 µm). This is sensible since redder colorsindicate cooler dust temperatures which naturally arise fartherfrom the central star. Unfortunately, this diagram is difficult touse when your target sample spans a large luminosity range. Thisis because the radius for dust at given temperature depends onthe root of central luminosity (see the size–luminosity diagramabove). Neither size–luminosity nor the size–color diagram cansimultaneously take into account both of these effects. Liu et al.(2007) also explored mid-IR size differences as a function ofsome SED parameters but his work also suffers from these samelimitations.

Despite these drawbacks, we have produced a similar diagramas a way to present our basic results. Figure 9(a) shows ourmeasured Gaussian FWHM size (in milliarcseconds) versus theIRAS 12–25 color. We chose to use the distance-independentangular size instead of the physical size here, just so that ourquantities did not depend on the uncertain distance estimates.Within each grouping, we do see some correlations. The HerbigAe stars do show a correlation, with the striking result that nearlyall the objects with colors bluer than 0.25 were unresolved byour survey. The embedded objects also might show a weak

correlation. Just as the diagram in Leinert et al. (2004), this size–color diagram does not conserve a target’s position if you takethe same disk temperature profile and change the luminosity.We only present this data to illustrate some trends within thesample, but caution against further use of this sort of diagramin subsequent work.

In order to overcome the luminosity and distance dependen-cies for this sort of diagram, we introduce a new diagram inFigure 9(b). Here, plot the luminosity-normalized ring radiusversus IRAS 12–25 color. This is defined as the physical ringradius (AU) divided by the square-root of luminosity (in so-lar luminosities). This quantity is both independent of distanceand also naturally accounts for the expected scaling of emissionscale with luminosity (for similar disk temperature profiles).Again, we see a significant correlation for the Herbig Ae stars,although other objects appear to scatter in this diagram. Thelocation of each YSO system in this diagram is diagnostic ofthe disk flaring and temperature profiles and this diagram willprove useful for future work.

In the next paper in this series, we will be carrying out aradiative transfer study of YSO disks in order to understandthe scatter in the size–luminosity diagram and how the de-viations correlate with observables such as stellar luminosityand the [12]–[25] µm color. This new work will motivate im-proved diagrams to be used for plotting basic observational dataand will explore the potential for using luminosity-normalizedsizes (radius AU/sqrt luminosity) and surface brightnessrelations.

5. SUMMARY

We have presented a large survey of mid-infrared sizes ofYSOs using the Keck telescope. We can reliably resolve diskswith size scales down to 35 mas (5 AU at 140 pc), roughly8× smaller than the formal diffraction limit. We have overcomethe traditional calibration problems that have plagued previous

10-1 100 101 102 103 104 105 106 107

luminosity (L⊙)

10-2

10-1

100

101

102

half-lig

ht ra

diu

s (A

U)

T = 90

0 KT =

250 K

transition disk

debris disk

Fig Mid-infrared size-luminosity relation: Monnier+ (2009) vs. MIDI

9 / 23


Near-IR vs. Mid-IR

Millan-Gabet+ (2007) Our MIDI sample

100 102 104 106

Lstar + Laccretion (Lsolar)

0.01

0.10

1.00

10.00

100.00

Rin

g R

adiu

s (A

U)

HAe ObjectsHBe ObjectsTTS Objects

Sublimation radius: direct dust heating, backwarmingSublimation radius: direct dust heating, no backwarmingSublimation radius: standard model

1000K

1500K1000K

1500K

1000K

1500Koptically−thin gas

(puffed−up

optically−thickgas

gas + dust

gas + dust

"Standard" Disk Model − oblique disk heating

Dust SublimationRadius Rs(magnetospheric

accretion?)

(magnetosphericaccretion?)

inner wall?)

flared disk surfaceDust Sublimation

flared disk surface

Star

Star

Radius Rs

Direct heating of inner dust disk

Fig. 2.— Measured sizes of HAeBe and T Tauri objectsvs. central luminosity (stellar+ accretion shock); and comparison withsublimation radii for dust directly heated by the central luminosity (solid and dashed lines) and for the oblique heating implied by theclassical models (dotted line). A schematic representation of the key features of inner disk structure in these two classes of models isalso shown. Data for HAeBe objects are from: IOTA (Millan-Gabet et al., 2001), Keck aperture masking (Danchi et al., 2001,Tuthillet al., 2001), PTI (Eisner et al., 2004) and KI (Monnier et al., 2005a). Data for T Tauri objects are from PTI (Akeson et al., 2005b) andKI (Akeson et al., 2005a,Eisner et al., 2005). For clarity, for objects observed at more than one facility, we include only the most recentmeasurement.

bution from scattered light originating from these relativelysmall spatial scales is negligible, allowing to consider fewermodel components and physical processes.

3.1 Disk Sizes and Structure

The geometry of proto-planetary disks is of great im-portance for a better understanding of the processes of diskevolution, dissipation and planet formation. The composi-tion of the dust in the upper disk layers plays a crucial rolein this respect, since the optical and UV photons of the cen-tral star, responsible for determining the local disk tempera-ture and thus the disk vertical scale-height, are absorbed bythese upper disk layer dust particles. The disk scale-heightplays a pivotal role in the process of dust aggregation andsettling, and therefore impacts the global disk evolution.Inaddition, the formation of larger bodies in the disk can leadto structure, such as density waves and gaps.

Spatially resolved observations of disks in the MIR us-ing single telescopes have been limited to probing relativelylarge scale (100s AU) thermal emission in evolved disksaround main-sequence stars (e.g.,Koerner et al., 1998;Jayawardhana, 1998) or scattering by younger systems(e.g., McCabe et al., 2003). Single-aperture interfero-metric techniques (e.g., nulling,Hinz et al., 2001; Liu etal., 2003, 2005 or segment-tilting,Monnier et al., 2004b)as well as observations at theInfrared Spatial Interferome-ter (ISI, Tuthill et al., 2002) have succeeded in resolving thebrightest young systems on smaller scales for the first time,establishing MIR sizes of∼ 10s AU. Although the observedsamples are still small, and thus far concern mostly HAeobjects, these initial measurements have already provided

a number of interesting, albeit puzzling results: MIR char-acteristic sizes may not always be well predicted by simple(flared) disk models (some objects are “under-sized” com-pared to these predictions, opposite to the NIR result), andmay not always connect in a straightforward way with mea-surements at shorter (NIR) and longer (mm) wavelengths.

These pioneering efforts were however limited in res-olution (for the single-aperture techniques) or sensitivity(for the heterodyne ISI), and probing the MIR emission onsmall scales has only recently become possible with the ad-vent of the new-generation long baseline MIR instruments(VLTI/MIDI, Leinert et al., 2003; KI/Nuller, Serabyn etal., 2004).

Two main factors determine the interferometric signa-ture of a dusty protoplanetary disk in the MIR: The overallgeometry of the disk, and the composition of the dust in thedisk “atmosphere.” The most powerful way to distentanglethese two effects is via spectrally resolved interferometry,and this is what the new generation of instruments such asMIDI at the VLTI are capable of providing.

First results establishing the MIR sizes of young circum-stellar disks were obtained by VLTI/MIDI for a sample ofseven HAeBe objects (Leinert et al. 2004). The characteris-tic sizes measured, based on visibilities at the wavelengthof12.5µm, are in the range 1–10 AU. Moreover, as shown inFig. 4, they are found to correlate with the IRAS [12]–[25]color, with redder objects having larger MIR sizes. Theseobservations lend additional support to the SED-based clas-sification ofMeeus et al. (2001) andDullemond and Do-minik (2004), whereby redder sources correspond to diskswith a larger degree of flaring, which therefore also emit

7

10-1 100 101 102 103 104 105 106 107

luminosity (L⊙)

10-2

10-1

100

101

102

half-lig

ht ra

diu

s (A

U)

T = 90

0 KT =

250 K

transition disk

debris disk

tight relation loose relationPhysics: dust sublimation Physics: flaring + gaps?

10 / 23


Flaring

Flaring disk intercepts more stellar flux

Mechanism:#(small grains) ↑ =⇒ flaring ↑

Observational appearance of disk:

1 larger

2 redder

⇒ small, large

Fig Strong vs. weak flaring

11 / 23


Color: the MIDI sample

−1.5 −1.0 −0.5 0.0 0.5 1.0 1.5 2.0MIDI 8 − 13

10-2

10-1

100

101R/L1/2 (

AU

L−1/2

⊙)

51

Op

hA

B_A

ur

AK

_Sco

AS

22

0 CQ

_Tau

CR

_Ch

a

CV

_Ch

a DI_

Ch

a

DR

_Tau

Eli

as2

-24

Eli

as2

-28

HD

31

64

8

HD

36

11

2

HD

36

91

7

HD

38

12

0

HD

72

10

6

HD

87

64

3

HD

95

88

1

HD

97

04

8

HD

98

92

2HD

10

04

53

HD

10

05

46

HD

10

42

37

HD

13

53

44

B

HD

13

96

14

HD

14

25

27

HD

14

25

60H

D1

42

66

6

HD

14

30

06

HD

14

44

32

HD

14

46

68

HD

15

01

93

HD

16

32

96

HD

17

92

18

HD

25

94

31

R_C

rA

RY

_Tau

SU

_Au

r

T_C

rA

UX

_Ori

V8

92

Tau

V2

06

2O

ph

V2

12

9O

ph

WW

_Ch

a

Fig Correlation between normalized size and color: larger disks are redder

small

large

12 / 23


Distinguishing flaring from gaps?

Model populationGrid of radiative transfer models for (gapless) disks(varying dust mass Mdust, surf. dens. p, settling parameter α, halo mass Mhalo, particle size distr. q)

−1.5 −1.0 −0.5 0.0 0.5 1.0 1.5 2.0MIDI 8 − 13

10-2

10-1

100

101

R/L1/2 (

AU

L−1/2

⊙)

51

Op

hA

B_A

ur

AK

_Sco

AS

22

0 CQ

_Tau

CR

_Ch

a

CV

_Ch

a DI_

Ch

a

DR

_Tau

Eli

as2

-24

Eli

as2

-28

HD

31

64

8

HD

36

11

2

HD

36

91

7

HD

38

12

0

HD

72

10

6

HD

87

64

3

HD

95

88

1

HD

97

04

8

HD

98

92

2HD

10

04

53

HD

10

05

46

HD

10

42

37

HD

13

53

44

B

HD

13

96

14

HD

14

25

27

HD

14

25

60H

D1

42

66

6

HD

14

30

06

HD

14

44

32

HD

14

46

68

HD

15

01

93

HD

16

32

96

HD

17

92

18

HD

25

94

31

R_C

rA

RY

_Tau

SU

_Au

r

T_C

rA

UX

_Ori

V8

92

Tau

V2

06

2O

ph

V2

12

9O

ph

WW

_Ch

a

−1.5 −1.0 −0.5 0.0 0.5 1.0 1.5 2.0MIDI 8 − 13

10-2

10-1

100

101

R/L1/2 (

AU

L−1

/2

⊙)

q

q = -2.8

q = -3.2

q = -3.5

q = -3.8

q = -4.5

Fig Gapless disks explained by flaring (i.e., amount of small grains)

13 / 23


Distinguishing flaring from gaps?

Fig Correlation between normalized size and color: larger disks are redder

14 / 23


Near-IR vs. Mid-IRREVISITED

Millan-Gabet+ (2007) Our MIDI sample

100 102 104 106

Lstar + Laccretion (Lsolar)

0.01

0.10

1.00

10.00

100.00

Rin

g R

adiu

s (A

U)

HAe ObjectsHBe ObjectsTTS Objects

Sublimation radius: direct dust heating, backwarmingSublimation radius: direct dust heating, no backwarmingSublimation radius: standard model

1000K

1500K1000K

1500K

1000K

1500Koptically−thin gas

(puffed−up

optically−thickgas

gas + dust

gas + dust

"Standard" Disk Model − oblique disk heating

Dust SublimationRadius Rs(magnetospheric

accretion?)

(magnetosphericaccretion?)

inner wall?)

flared disk surfaceDust Sublimation

flared disk surface

Star

Star

Radius Rs

Direct heating of inner dust disk

Fig. 2.— Measured sizes of HAeBe and T Tauri objectsvs. central luminosity (stellar+ accretion shock); and comparison withsublimation radii for dust directly heated by the central luminosity (solid and dashed lines) and for the oblique heating implied by theclassical models (dotted line). A schematic representation of the key features of inner disk structure in these two classes of models isalso shown. Data for HAeBe objects are from: IOTA (Millan-Gabet et al., 2001), Keck aperture masking (Danchi et al., 2001,Tuthillet al., 2001), PTI (Eisner et al., 2004) and KI (Monnier et al., 2005a). Data for T Tauri objects are from PTI (Akeson et al., 2005b) andKI (Akeson et al., 2005a,Eisner et al., 2005). For clarity, for objects observed at more than one facility, we include only the most recentmeasurement.

bution from scattered light originating from these relativelysmall spatial scales is negligible, allowing to consider fewermodel components and physical processes.

3.1 Disk Sizes and Structure

The geometry of proto-planetary disks is of great im-portance for a better understanding of the processes of diskevolution, dissipation and planet formation. The composi-tion of the dust in the upper disk layers plays a crucial rolein this respect, since the optical and UV photons of the cen-tral star, responsible for determining the local disk tempera-ture and thus the disk vertical scale-height, are absorbed bythese upper disk layer dust particles. The disk scale-heightplays a pivotal role in the process of dust aggregation andsettling, and therefore impacts the global disk evolution.Inaddition, the formation of larger bodies in the disk can leadto structure, such as density waves and gaps.

Spatially resolved observations of disks in the MIR us-ing single telescopes have been limited to probing relativelylarge scale (100s AU) thermal emission in evolved disksaround main-sequence stars (e.g.,Koerner et al., 1998;Jayawardhana, 1998) or scattering by younger systems(e.g., McCabe et al., 2003). Single-aperture interfero-metric techniques (e.g., nulling,Hinz et al., 2001; Liu etal., 2003, 2005 or segment-tilting,Monnier et al., 2004b)as well as observations at theInfrared Spatial Interferome-ter (ISI, Tuthill et al., 2002) have succeeded in resolving thebrightest young systems on smaller scales for the first time,establishing MIR sizes of∼ 10s AU. Although the observedsamples are still small, and thus far concern mostly HAeobjects, these initial measurements have already provided

a number of interesting, albeit puzzling results: MIR char-acteristic sizes may not always be well predicted by simple(flared) disk models (some objects are “under-sized” com-pared to these predictions, opposite to the NIR result), andmay not always connect in a straightforward way with mea-surements at shorter (NIR) and longer (mm) wavelengths.

These pioneering efforts were however limited in res-olution (for the single-aperture techniques) or sensitivity(for the heterodyne ISI), and probing the MIR emission onsmall scales has only recently become possible with the ad-vent of the new-generation long baseline MIR instruments(VLTI/MIDI, Leinert et al., 2003; KI/Nuller, Serabyn etal., 2004).

Two main factors determine the interferometric signa-ture of a dusty protoplanetary disk in the MIR: The overallgeometry of the disk, and the composition of the dust in thedisk “atmosphere.” The most powerful way to distentanglethese two effects is via spectrally resolved interferometry,and this is what the new generation of instruments such asMIDI at the VLTI are capable of providing.

First results establishing the MIR sizes of young circum-stellar disks were obtained by VLTI/MIDI for a sample ofseven HAeBe objects (Leinert et al. 2004). The characteris-tic sizes measured, based on visibilities at the wavelengthof12.5µm, are in the range 1–10 AU. Moreover, as shown inFig. 4, they are found to correlate with the IRAS [12]–[25]color, with redder objects having larger MIR sizes. Theseobservations lend additional support to the SED-based clas-sification ofMeeus et al. (2001) andDullemond and Do-minik (2004), whereby redder sources correspond to diskswith a larger degree of flaring, which therefore also emit

7

10-1 100 101 102 103 104 105 106 107

luminosity (L⊙)

10-2

10-1

100

101

102

half-lig

ht ra

diu

s (A

U)

T = 90

0 KT =

250 K

transition disk

debris disk

tight relation loose relationPhysics: dust sublimation Physics: flaring + gaps

15 / 23


Flaring: correlation with spectra?

Physics of flaring: #(small grains) ↑ =⇒ flaring ↑

Expected: correlation between disk size and spectra(= probe for small grains)

F. Przygodda et al.: Evidence for grain growth in T Tauri disks L45

1

1.5

2

2.5

3

3.5

8 9 10 11 12 13

AK Sco

1

1.5

2

2.5

3

3.5

8 9 10 11 12 13

Haro 1-16

1

1.5

2

2.5

3

3.5

8 9 10 11 12 13

GW Ori

1

1.5

2

2.5

3

3.5

8 9 10 11 12 13

CR Cha

1

1.5

2

2.5

3

3.5

8 9 10 11 12 13

TW Hya

1

1.5

2

2.5

3

3.5

8 9 10 11 12 13

Glass I

1

1.5

2

2.5

3

3.5

8 9 10 11 12 13

SU Aur

1

1.5

2

2.5

3

3.5

8 9 10 11 12 13

GG Tau

1

1.5

2

2.5

3

3.5

8 9 10 11 12 13

WW Cha

1

1.5

2

2.5

3

3.5

8 9 10 11 12 13

AS 205 SW component

1

1.5

2

2.5

3

3.5

8 9 10 11 12 13

S CrA SE component

1

1.5

2

2.5

3

3.5

8 9 10 11 12 13

DR Tau

1

1.5

2

2.5

3

3.5

8 9 10 11 12 13

HBC 639

1

1.5

2

2.5

3

3.5

8 9 10 11 12 13

AS 205 NE component

1

1.5

2

2.5

3

3.5

8 9 10 11 12 13

RU Lup

1

1.5

2

2.5

3

3.5

8 9 10 11 12 13

S CrA NW component

lambda (µm)

Fig. 2. Continuum normalized spectra of our sample ordered by the strength of the silicate feature. The shape of the feature is showing acorrelation with the strength. Stronger features have a triangular shape with a pronounced peak near 9.8µm while weaker features are moreplateau-like. The gap from 9.15 to 9.65µm in most of the spectra is caused by a broken channel of the TIMMI2 detector.

0

0.5

1

1.5

2

2.5

3

3.5

4

8 9 10 11 12 13

abso

rptio

n co

eff.

(103 c

m2 g-1

)

lambda (µm)

olivine 0.1 µm grainsolivine 2.0 µm grains

Fig. 3. Theoretical spectra of olivine dust with grain sizes of 0.1µmand 2.0µm (from Bouwman et al. 2001). The points at 9.8µm and11.3µm are visualizing the dependence of the flux ratio at these wave-lengths on the grain size.

– the mid-infrared absorption coefficient of larger amorphousolivine dust grains shows a plateau-like shape in the rangefrom 9.5 to 12µm.

Assuming a dust temperature of around 300 K, the blackbodythermal emission will not vary very much over our wavelengthband, and the silicate emission profiles would have a shapequite close to the shape of the absorption coefficients. Since

the absorption cross section per unit mass of large grains isingeneral smaller than that of small grains of similar shape, theemission of large grains is in general expected to be weakerthan that of small grains.

Our observations suggest that most observed mid-infraredspectra of young low or intermediate mass stars can be char-acterised simply by emission corresponding to the two grainsizes covered in Fig. 4. We note that the ratio of absorption co-efficients at the wavelengths of 11.3µm and 9.8µm as functionof grain size varies from 0.43 (for 0.1µm grains) to about 1.0(2.0 µm grains). Therefore, the ratio of the (continuum sub-tracted) flux at 11.3µm over that at 9.8µm should be a mea-sure of the grain size. In Fig. 4 we plotted this ratio againstthe strength of the silicate feature. The distribution of the datapoints shows a clear correlation. Strong silicate featureshave11.3/9.8 flux ratios down to 0.53 (for CR Cha), while weakfeatures exhibit higher ratios up to about 1.0 (in case of theS CrA NW-component even 1.42). The remaining points aredistributed over the range between these two extremes. The er-ror bars quoted in Fig. 4 take into account the signal to noiseof the observations, and the uncertainty in the continuum es-timate. A weighted linear fit to the data points with the rela-tion y = a − bx, where x is the strength of the feature andy the flux ratio, gives the coefficients:a = 1.11 ± 0.11 andb = −0.16± 0.04. A very similar correlation between strength

Letterto

theEditor

Fig Emission of sub-micron and micron-sized silicates (Przygodda+ 2003)

16 / 23


Flaring vs. gaps?

Fig Correlation between normalized size and grain processing?

(silicate-feature sources only)

small grains large grains

17 / 23


Flaring vs. gaps?

Fig Correlation between normalized size and grain processing?

(silicate-feature sources only)

small grains large grains

18 / 23


First results: 2. Towards a family picture

Two-layer model, homogeneous disk

Physics:

1 inclination i, position angle PA

2 T (R) ∝ R−q, with Rin = Rsub and Rout = 100 AU

3 silicate dust species + carbonaceous grains: κi

19 / 23


HD 163296

10 AU

8 9 10 11 12 13wave (µm)

0

5

10

15

20

25

30

flux (Jy)

0.0m, 0.0deg

8 9 10 11 12 13wave (µm)

0

5

10

15

20

25

30

flux (Jy)

99.4m, 17.7deg

8 9 10 11 12 13wave (µm)

0

5

10

15

20

25

30

flux (Jy)

16.8m, 92.7deg

8 9 10 11 12 13wave (µm)

0

5

10

15

20

25

30

flux (Jy)

16.4m, 93.1deg

8 9 10 11 12 13wave (µm)

0

5

10

15

20

25

30

flux (Jy)

14.2m, 81.3deg

8 9 10 11 12 13wave (µm)

0

5

10

15

20

25

30

flux (Jy)

12.8m, 84.4deg

8 9 10 11 12 13wave (µm)

0

5

10

15

20

25

30

flux (Jy)

31.4m, 66.1deg

8 9 10 11 12 13wave (µm)

0

5

10

15

20

25

30

flux (Jy)

31.7m, 67.8deg

8 9 10 11 12 13wave (µm)

0

5

10

15

20

25

30

flux (Jy)

30.5m, 78.1deg

8 9 10 11 12 13wave (µm)

0

5

10

15

20

25

30

flux (Jy)

28.9m, 80.7deg

8 9 10 11 12 13wave (µm)

0

5

10

15

20

25

30

flux (Jy)

28.2m, 81.7deg

8 9 10 11 12 13wave (µm)

0

5

10

15

20

25

30

flux (Jy)

74.7m, 104.2deg

8 9 10 11 12 13wave (µm)

0

5

10

15

20

25

30

flux (Jy)

75.5m, 104.4deg

8 9 10 11 12 13wave (µm)

0

5

10

15

20

25

30

flux (Jy)

43.7m, 57.8deg

8 9 10 11 12 13wave (µm)

0

5

10

15

20

25

30

flux (Jy)

44.5m, 59.5deg

8 9 10 11 12 13wave (µm)

0

5

10

15

20

25

30

flux (Jy)

40.7m, 146.3deg

8 9 10 11 12 13wave (µm)

0

5

10

15

20

25

30

flux (Jy)

40.6m, 152.1deg

8 9 10 11 12 13wave (µm)

0

5

10

15

20

25

30

flux (Jy)

40.1m, 158.7deg

8 9 10 11 12 13wave (µm)

0

5

10

15

20

25

30

flux (Jy)

60.6m, 122.2deg

8 9 10 11 12 13wave (µm)

0

5

10

15

20

25

30

flux (Jy)

59.3m, 124.1deg

HD 163296: model image (left), fit to correlated fluxes (right)20 / 23


HD 139614

10 AU

8 9 10 11 12 13wave (µm)

0

1

2

3

4

5

6

flux (Jy)

0.0m, 0.0deg

8 9 10 11 12 13wave (µm)

0

1

2

3

4

5

6

flux (Jy)

49.8m, 81.7deg

8 9 10 11 12 13wave (µm)

0

1

2

3

4

5

6

flux (Jy)

54.8m, 90.0deg

8 9 10 11 12 13wave (µm)

0

1

2

3

4

5

6

flux (Jy)

61.2m, 105.1deg

8 9 10 11 12 13wave (µm)

0

1

2

3

4

5

6

flux (Jy)

52.4m, 34.5deg

8 9 10 11 12 13wave (µm)

0

1

2

3

4

5

6

flux (Jy)

56.1m, 2.1deg

8 9 10 11 12 13wave (µm)

0

1

2

3

4

5

6

flux (Jy)

55.6m, 16.4deg

HD 139614: model image (left), fit to correlated fluxes (right)(cf. Matter+ 2014)

21 / 23


A family picture?

10 AU 10 AU 10 AU 10 AU 10 AU

10 AU 10 AU 10 AU 10 AU 10 AU

Preliminary “images”: HD 163296, HD 139614, 51 Oph, HD 104237,

RY Tau, HD 144668, HD 98922, HD 36112, HD 179218, HD 142560

22 / 23


Conclusions

MIDI: structure (and composition) of protoplanetary disks

The MIDI-Herbig project: statistics!

Preliminary results:

1 size-luminosity diagram for full sample of objects:flaring vs. gaps?

2 first model images

? ? ?

23 / 23

as seen with MIDI - Max Planck Society · 2014. 5. 14. · IntroThe MIDI-Herbig projectFirst...

Documents

Transcript of as seen with MIDI - Max Planck Society · 2014. 5. 14. · IntroThe MIDI-Herbig projectFirst...