July 8, 2008Polarized Fluorescence in RNe - AstroPol081 Atomic fluorescence and prospects for observing magnetic geometry using magnetic realignment of.

July 8, 2008 Polarized Fluorescence in RNe - AstroPol08 1

Atomic fluorescence and prospects for observing magnetic geometry

using magnetic realignment of atomic ground states

• Magnetic realignment - Observer's perspective• SALT fluorescence pilot observations: NGC2023• A “Spreadsheet” Model

– Which Ions

– Comparison to observations

– Planning polarization observations

• Future/ Observations


Magnetic Realignment

• Linear polarization of atomic resonance/ fluorescence lines – anisotropic UV pumping produces anisotropic angular momentum

distribution ("alignment") of ground state if photon rate > collisions (i.e., certainly within 1-10 pc of OB*)

– presence of magnetic field alters alignment if Larmor freq > photon rate • ISM: B > ~0.1 μGauss: effect is “saturated”, field geometry only• CSM: B > 10-104 μGauss: could depend on field strength also

• Unique polarization signatures:– non-zero net polarization of IS/ CS absorption lines– distortion of position angles of scattered emission lines

• Depends on: 3D geometry of magnetic field, ion ground state configuration and pumping (Yan & Lazarian 2006 - 2008)

• Potentially more powerful than– 21 cm Zeeman: sensitive to weaker fields; works in hot gas– dust alignment: sensitive to 3D geometry, gas props and velocity


How to Observe it?I. Absorption

• Ions: need at least 3 fine states in ground level (J ≥ 1)

– Neutral: NI, OI, SII, FeII

• Resolution. For sensitivity, resolve IS lines (R > 20,000)

• Wavelengths: almost entirely in the FUV (except TiII, FeI - really complicated)

• Tough, but will be trying it with Far Ultraviolet SpectroPolarimeter on θ1 and θ2 Ori through "Orion Veil" θ2

θ1


How to Observe it?II. Emission

• Ions: need at least 3 fine/ hyperfine states in ground level (F ≥ 1)– InterPlan, PDR: NaI, KI

– Neutral: NI, OI, SII, FeII, AlII

• Resolution. For sensitivity, against dust continuum (R > 5,000)

• Wavelengths: – Resonance: UV, except NaI, KI

– Fluorescence: UV/Vis/NIR

• Signal: position angle rotation from reflection polarization (centrosymmetric)

• Can observe this with ground-based high-resolution spectropolarimeter!

Pmax = 19.1%O I


Pilot Observation: Fluorescence in Reflection

Nebulae• OI, NI fluorescence previously seen only in

HII, PNe: weak lines, with many excitation processes

• Better: in RNe, is only excitation process; but need to verify/ model, prepare for spectropolarimetry

• Robert Stobie Spectrograph on SALT 11m:– NGC2023 RN, HD37903 B1.5V central star

– 0.6”x8’ slit, 1st order, R = 7500 - 9500

– ~1000 Ang coverage blue, yellow, red

• Many fluorescence lines found! Inner 2’ of nebula, within PDR:

NI OI SiII TiII CrII FeII ?

10 5 3 1? 1? 17 6


Predicted Ions

• Which Ions?– Neutral medium: principle ion with IP < 13.6 eV– Abundance/H > 10-10

– 1st Resonance < 13.6 eV

• 8 with LS Coupling (primary, secondary, alignable):

Atom Ion Res Vis1 Vis2

C II 2 0 0

N I 4 0 18

O I >12 11 >2

Mg II 7 10 4

Al II 2 2 0

Si II 16 4 5

P II 15 0 0

S II 5 0 0

• 6 with Non-LS coupling: ArI, TiII, CrII, MnII, FeII, NiII

• We do indeed see these!


Predicted Equivalent Widths

• Model scattered line / visible continuum = Equivalent Width

• Observed lines: get generally correct EW– one OI may be confused with FeII– MgII not seen. depletion?

• Predict more lines for future NIR instrument• Predict more Vis lines: AlII, strong MgII

Primary Fluorescence

0.0

0.1

1

10

100

3000 5000 7000 9000 11000 13000 15000 17000

N I

O I

S i II

A l II

M g II

`

Pre

dic

ted

EW

(A

ng

)

R SS N IRR SS Vis

Secondary FluorescenceN I

O I

S i II

A l II

M g II

N G C 2023 F luorescence Predictions

0.01

0.1

1

10

0.001 0.01 0.1 1 10Predicted EW

Ob

serv

ed

EW

(A

ng

)

N I

O I

S i II

M g II

F illed - P rim aryO pen - Secondary

A (m ag)V 0.20H /H2

0.000b(km /s) 5.0log Ly -11.2


Expected Polarization Signals

• Reflection polarization– from unaligned ground state

(“thermalized”)– like electron scattering (100%

at 90°) times “polarizability” E1: depends only on J of levels

– position angle perp or parallel to radius vector

– pol depends only on scattering angle: deduce geometry

PA magnified 30x Pmax = 18%Al II 8643

• Alignment polarization– for alignable ions,

ground state aligned by pumping via all UV resonance transitions

– changes pol and PA depending on scattering angle, 3D magnetic field orientation

– recognizable signal is Uperp = p sin 2ΔPA. Use map of this to deduce mag field orientation


Diagnostic Diagram

• Plot Realignment sensitivity vs polarizability to select lines to map

• Geometry, calibration– Mg II 9246: E1 = 50%! Use

to deduce scattering angle map

– MgII 9221, Si 5981: E1 = 0. Use to measure foreground interstellar polarization

• Magnetic field determination– OI 7997, Al 8643: Uperp =>

magnetic field map

O I (D ->)

M g II (3 /2->3/2)

A l II (->S)

S i II (3 /2->1/2)S i II (3 /2->3/2)

M g II (3 /2->1/2)M g II (3 /2->5/2)

A l II (->D )O I (S ->)

M g II (1 /2->)S i II (S ->)

-5

0

5

10

-60 -30 0 30 60Polarizability E (%)1

Alig

nm

ent

U (

%)

per

p


Possible Hitch: Optical Depth

• Effect of optical depth of "trapped" UV transitions (τ >> 1):– Fluorescence intensity keeps growing with τ

– depolarization due to more isotropic diffuse radiation

• Good news: many fluorescent excitors never trapped: τeff < 1/(escape prob): retain reflection polarization (small symbols)

• Not as good: many alignable ions have trapped pumping lines: alignment polarization signal depolarized (large symbols)

• Remedies:– look at thinner nebulae (signal still good)

– look at FeII: not trapped

NGC2023


Summary/ Future

• Atomic Fluorescence lines seen for the first time in a Reflection nebula: NI, OI, SiII, FeII– Intensities consistent with simple model– Important to understand pumping in optically thick nebula

• Expect linear polarization signal– Reflection polarization (polarizability) should be easily

observable– Magnetic realignment polarization distortion observable with

OI, AlII, but may be reduced by trapped pumping lines– Best bet: FeII

• Future NGC2023 – Spectroscopy: look for MgII, AlII lines, – Slit spectropolarimetry: verify best polarized lines– Fabry-Perot spectropolarimetry: field map– Model with CLOUDY, do realignment calculations for FeII– Move on to thinner nebulae, HII regions...

Orion Neb Hα

IC2118

“Witch’s Head”


Extra Slides


What Gas is This?

• Line signal goes to zero 1’ from illuminating star HD37903

• peaks sharply in clump 30” N• Lies almost entirely inside H2 emission in

well studied PhotoDissociation Region• New probe of this warm neutral material:

previous absorption studies dominated by cold foreground material– velocity probe– temperature probe– magnetic field probe, too warm & thick for

HI Zeeman


H /H = 0.1A = 0.13

2

V

0

1

Model: Illumination

• Fluorescence excited by FUV: 912 – 1200 Ang

• Unextincted HD37903: use FUSE HD121300

• Assume physically thin shell, standard dust/ gas, standard extinction, variable AV

• H2 absorption important: allow variable H2/H

S i II

A l II

M g II

O I

N I

900 1000 1100 1200W avelength (Ang)

Flu

x


NI, AlII


MgII, Si II

July 8, 2008Polarized Fluorescence in RNe - AstroPol081 Atomic fluorescence and prospects for observing magnetic geometry using magnetic realignment of.

Documents

atomic fluorescence

ni fluorescence

velocitypolarized fluorescence

redmany fluorescence

lines r

rne astropol08how

rne astropol08expect

magnetic geometry