RHESSI and the Solar Flare Xray Spectrum Ken

RHESSI and the Solar Flare X-ray Spectrum Ken Phillips Presentation at Wroclaw Workshop “X-ray spectroscopy and plasma diagnostics from the RESIK, RHESSI and SPIRIT Instruments “, December 6— 8, 2005

Reuven Ramaty High Energy Solar Spectroscopic Imager (RHESSI)

RHESSI Launch from Pegasus Launch of RHESSI: Feb. 5, 2002 Pegasus rocket carrying RHESSI NASA plane

RHESSI Spacecraft Concept S/C Rotates SUN

RHESSI’s Energy Range • RHESSI observes soft and hard X-ray emission up to gamma-rays: 4 ke. V up to 17 Me. V • Conversion is λ(Å) = 12. 4 / E (ke. V)

RHESSI 10— 25 ke. V images of near-limb flare Lin. Hui Sui & Holman 2003 Solar Limb Spatial resolution in these images 7 arcsec (can be as good as 2 arcsec)

Holman et al. (2002) -- HXR and Gamma-ray images -- RHESSI, 2002 July 23 flare SOHO/MDI magnetogram HXR emission Gamma rays

TRACE and RHESSI Flare Images

RHESSI as a Spectrometer: Spectral Resolution

• Spectral resolution best for RHESSI . Detectors 3 and 4: 1 ke. V at ~ 7 ke. V or ~0. 25Å at 1. 9Å. • Spectral resolution “broad-band”, not like crystal spectrometer resolution. • RHESSI sees Fe XXV lines and satellites as a single “bump” on continuum. • However, RHESSI does see the continuum (whereas crystal spectrometers usually affected by crystal fluorescence).

RHESSI flare spectrum

Flare X-ray Spectra • Electrons accelerated at reconnection site • travel from loop to chromosphere, producing X-ray emission with nonthermal spectrum (hv > 20 ke. V). Material in previously empty loop filled with heated plasma (T~20 MK) that produces a thermal spectrum (hv < 20 ke. V) consisting of lines and continuum.

Two Line features in RHESSI spectra 6. 7 ke. V or “Fe line feature” – Fe XXV lines – Fe XVIII – Fe XXIV satellites – Fe XXVI lines at higher T (>80 MK) – Seen with crystal spectrometers 8 ke. V or “Fe/Ni line feature” – More highly excited Fe XXV, Fe XXVI lines – Ni XXVII, Ni XXVIII lines & satellites – Never seen with crystal spectrometers

Fe and Fe/Ni line features: RHESSI observations So RHESSI the first instrument (apart from NEAR-PIN) to see the 8 ke. V feature By analyzing the 2 line features, one can diagnose the hottest part of the flare thermal plasma

Chianti Spectrum, T=20 MK Ca XIX Fe XXVI Fe XXV +satellites Ni XXVII + sats Fe edge

Chianti spectra + spectra smoothed to RHESSI resolution

SMM/BCS Fe line flare spectra Fluorescence background

Synthetic (CHIANTI) spectra, solar plasma (8 -33 MK) RHESSI (1 ke. V) resolution assumed Chianti: coronal abund. + Mazzotta et al. ion. eq.

Lines making up the `Fe-line’ (6. 7 ke. V) feature

Lines making up the `Fe/Ni’ (8 ke. V) feature

Solar flare spectrum, NEARPIN and Yohkoh/BCS S XV BCS NEAR-PIN: R. Starr Yohkoh: S, Ca, Fe channels Ca XIX NEAR-PIN Fe line (6. 7 ke. V) Fe/Ni line (8 ke. V)

Element (e. g. Fe) Abundances • With measurement of line and continuum flux, we can measure element abundances • Thus, with 6. 7 ke. V (“Fe line”) feature, we can measure Fe/H. • Use equivalent width of line to do this.

Emission Line Equivalent Width Definition Energy (ke. V)

Equivalent Width of the Fe line Based on Chianti v. 5. 1, coronal abundances and Mazzotta et al. ion fractions (This plot is different from Phillips 2004)

Attenuator states RESIK in 3 rd order

Theoretical curve (Phillips 2004) based on coronal Fe abundance and Mazzotta et al. ion fractions

F e Theoretical curve

Equivalent Width Plots • April 23, 2003 flare is `good’ example – plenty of not-so-good examples! • There are problems associated with high count rates in the A 0 attenuator state and in the A 3 attenuator state – these are not understood at present. • Cristina will show more examples…. • Tentative conclusion: flare Fe abundance ~ coronal Fe abundance (Feldman & Laming 2000)

Fe XXIV/Fe XXV Ion Fractions wrong? • In CHIANTI analysis, we used Mazzotta et al. • • (1998) ion fractions – maybe slightly wrong? No measured rates for these ions, so everything relies on calculations There are some ‘measured’ rates from SMM/BCS flare spectra: Fe XXIV satellites (j and q) / Fe XXV w line ratios (Antonucci et al. )

RHESSI and CORONAS-F/RESIK Solar Flare Spectra • Both RHESSI and RESIK spectrometers carefully calibrated – estimated uncertainties ~20% • Useful to compare spectra for same times in flares – agreement very good. • RESIK 1 st-order spectra agree with the low -energy end of RHESSI spectra.

Plotting spectra from 7 RHESSI detectors + RESIK

Summary and Conclusions • RHESSI a `broad-band’ spectrometer, sees both thermal and nonthermal spectra of solar flares (sp. res. 1 ke. V). • RHESSI sees two line features, at 6. 7 ke. V (Fe) and 8 ke. V (Fe/Ni). • Line equivalent widths mostly indicate coronal abundance of Fe. • Possibly ionization fractions of Fe XXV need revision.