energy solar phenomena: MSFC: B. Ramsey, S. Bongiorno GSFC: D. - - PowerPoint PPT Presentation

X-ray diagnostics of high energy solar phenomena: from the RHESSI spacecraft to the FOXSI sounding rocket

Sophie Musset University of Minnesota

Plasma Astrophysics Seminar KU Leuven 10 May 2019

Collaborators: Lindsay Glesener (University of Minnesota) Nicole Vilmer (LESIA, Observatoire de Paris) Eduard Kontar (University of Glasgow) FOXSI-3 Team UMN: Lindsay Glesener (PI), P .S. Athiray, S. Musset, J. Vievering UCB: S. Courtade, J.-C. Buitrago Casas,

• S. Krucker, G. Dalton, P

. Turin MSFC: B. Ramsey, S. Bongiorno GSFC: D. Ryan, S. Christe Kavli IPMU: T . Takahashi, K. Furukawa ISAS: S. Watanabe NAOJ: N. Narukage Nagoya Univ.: S. Ishikawa, I. Mitsuishi Tokyo Univ. of Science: K. Hagino

Solar flare: Sudden release of magnetic energy

15 fev 2011 304 Å 335 Å (extreme ultraviolet)

Heating Particle acceleration

21 aug 2002 195 Å 6-8 keV 25-80 keV

X-rays

1

Magnetic field lines Acceleration site

Photosphere

e- e-

X-ray diagnostic of solar flares

2

Magnetic reconnection & Magnetic energy dissipation in the corona Plasma heating Particle acceleration

Krucker et al. 2008

Magnetic field lines Acceleration site

X-ray diagnostic of solar flares

2

Magnetic reconnection & Magnetic energy dissipation in the corona Plasma heating Particle acceleration X-ray emission (bremsstrahlung)

Thermal emission (hot plasma) Non-thermal emission (accelerated electrons)

Krucker et al. 2008

Magnetic field lines Acceleration site

X-ray diagnostic of solar flares

2

Magnetic reconnection & Magnetic energy dissipation in the corona Plasma heating Particle acceleration X-ray emission (bremsstrahlung)

Thermal emission (hot plasma) Non-thermal emission (accelerated electrons)

Krucker et al. 2008

Magnetic field lines Acceleration site

Magnetic reconnection & Magnetic energy dissipation in the corona Plasma heating Particle acceleration X-ray emission (bremsstrahlung)

X-ray diagnostic of solar flares

Coronal HXR source (non-thermal emission): fainter

2

Outline

• Using X-ray imaging spectroscopy with RHESSI to constrain particle

transport in the solar corona

▪ Diffusive transport of energetic electrons in coronal loops

• Going further with focusing optics for solar X-ray diagnostics (FOXSI)

▪ Why do we need focusing optics for solar X-ray telescopes? ▪ The FOXSI sounding rocket

• Recent results from the FOXSI sounding rocket

▪ Micro-flare observations with FOXSI-2 ▪ FOXSI-3 flight on Sept 7 2018

3

Krucker et al. 2008

Magnetic field lines Acceleration site

Coronal HXR sources

From a photon spectrum to an electron spectrum

𝜹

𝒒𝒊𝒑𝒖𝒑𝒐 𝒕𝒒𝒇𝒅𝒖𝒔𝒗𝒏 4

Krucker et al. 2008

Magnetic field lines Acceleration site

Coronal HXR sources

From a photon spectrum to an electron spectrum Thin target approximation 𝜹 = 𝜺 + 𝟐

𝜹

𝒒𝒊𝒑𝒖𝒑𝒐 𝒕𝒒𝒇𝒅𝒖𝒔𝒗𝒏 4

Krucker et al. 2008

Magnetic field lines Acceleration site

Coronal HXR sources

From a photon spectrum to an electron spectrum Thin target approximation Thick target approximation 𝜹 = 𝜺 + 𝟐 𝜹 = 𝜺 − 𝟐

𝜹

𝒒𝒊𝒑𝒖𝒑𝒐 𝒕𝒒𝒇𝒅𝒖𝒔𝒗𝒏 4

Krucker et al. 2008

Magnetic field lines Acceleration site

Coronal HXR sources

From a photon spectrum to an electron spectrum Thin target approximation Thick target approximation 𝜹 = 𝜺 + 𝟐 𝜹 = 𝜺 − 𝟐 « Standard » collisional propagation model 𝜹𝒅𝒕 − 𝜹𝒈𝒒 = 𝟑 ሶ 𝑶𝒅𝒕 = ሶ 𝑶𝒈𝒒

𝜹

𝒒𝒊𝒑𝒖𝒑𝒐 𝒕𝒒𝒇𝒅𝒖𝒔𝒗𝒏 4

X-ray imaging spectroscopy with RHESSI

Battaglia & Benz (2006)

5

X-ray imaging spectroscopy with RHESSI

Battaglia & Benz (2006)

Photon spectral index

𝛿𝑀𝑈 − 𝛿𝐺𝑄 ≠ 2

Need additional mechanism to collisional transport 5

X-ray imaging spectroscopy with RHESSI

Battaglia & Benz (2006) Simoes & Kontar (2013)

Photon spectral index

𝛿𝑀𝑈 − 𝛿𝐺𝑄 ≠ 2

Electron spectral index

𝜀𝑀𝑈 − 𝜀𝐺𝑄 < 0

Ratio of electron rate

ሶ 𝑂𝑀𝑈 ሶ 𝑂𝐺𝑄 > 1

Need additional mechanism to collisional transport Need additional mechanism to collisional transport 5

Combination of radio and X-ray diagnostics

X-rays: Bremsstrahlung emission emitted by electrons of a few tens of keV Radio: Gyrosynchrotron emission emitted by electrons of a few hundred of keV

X-ray and radio diagnostics offer the possibility to study energetic electron transport in two different energy ranges

e- e- p+ 6

Gyrosynchrotron emission (electrons with a few hundred of keV) X-ray emission (electrons < 100 keV)

Kuznetsov & Kontar (2015)

Combination of radio and X-ray diagnostics

X-rays: Bremsstrahlung emission emitted by electrons of a few tens of keV Radio: Gyrosynchrotron emission emitted by electrons of a few hundred of keV

X-ray and radio diagnostics offer the possibility to study energetic electron transport in two different energy ranges

e- e- p+ 6

Gyrosynchrotron emission (electrons with a few hundred of keV) X-ray emission (electrons < 100 keV)

Kuznetsov & Kontar (2015)

Radio diagnostics of May 21 2004 flare

Spatial distribution of the density of energetic electrons with E > 60 keV

Energetic electrons are trapped in the corona, near the looptop

6

Musset et al. (2018)

X-ray imaging spectroscopy

Count flux [counts/s/cm2/keV] Energy [keV] 10 100 102 100 10-2 12-25 keV 25-50 keV 50-100 keV

X-ray imaging spectroscopy

Musset et al. (2018)

7

Musset et al. (2018)

X-ray imaging spectroscopy

Count flux [counts/s/cm2/keV] Energy [keV] 10 100 102 100 10-2 𝜀 = 5.2 ± 0.4 𝜀 = 4.4 ± 0.2 𝜀 = 4.2 ± 0.2 12-25 keV 25-50 keV 50-100 keV Coronal ambiant density 𝑜 = 1.2 ± 0.2 × 1011 𝑑𝑛−3

ሶ 𝑂 = 0.12 ± 0.03 × 1035 s−1 ሶ 𝑂 = 0.06 ± 0.02 × 1035 s−1

ሶ 𝑶𝑴𝑼 = 0.4 ± 0.2 × 1035 s−1

ሶ 𝑶𝑴𝑼 ሶ 𝑶𝑮𝑸 = 2.2 → Electrons are trapped

in the coronal part of the loop

X-ray imaging spectroscopy

Musset et al. (2018)

7

Energetic electron density above 25 keV (from X-rays) Energetic electron density above 60 keV

From radio (Kuznetsov & Kontar 2015) From X-rays

𝑜𝑐,𝑀𝑈

25

𝑜𝑐,𝐺𝑄

25

~1.6 𝑏𝑜𝑒 3.8 𝑜𝑐,𝑀𝑈

60

𝑜𝑐,𝐺𝑄

60

~7.7 𝑏𝑜𝑒 9 Distribution deduced from gyrosynchrotron emission is more peaked than the distribution deduced from X-rays Is trapping energy-dependent?

Musset et al. (2018)

Trapping of energetic electrons in the corona

8

1 𝑤 𝜖 𝜖𝑨 𝐸𝑨𝑨

(𝑈) 𝜖𝐺

𝜖𝑨 = 𝜖 𝜖𝐹 𝑒𝐹 𝑒𝑦 𝐺 + 𝐺0𝑇(𝑨) Diffusion Collisions Source 𝐸𝑨𝑨

(𝑈) = λ𝑤

3 λ : sca cattering mean free path

Strong pitch angle scattering due to small scale magnetic fluctuations ➔ diffusive transport of energetic electrons

Kontar et al. (2014)

Diffusive transport of energetic electrons

9

1 𝑤 𝜖 𝜖𝑨 𝐸𝑨𝑨

(𝑈) 𝜖𝐺

𝜖𝑨 = 𝜖 𝜖𝐹 𝑒𝐹 𝑒𝑦 𝐺 + 𝐺0𝑇(𝑨) Diffusion Collisions Source 𝐺𝐸 𝐹, 𝑨 = 𝐹 𝐿𝒐𝟏 න

𝐹 ∞

𝑒𝐹′ 𝑮𝟏 𝐹′ 4𝜌𝒃 𝐹′2 − 𝐹2 + 2𝒆2 exp − 𝑨2 4𝒃 𝐹′2 − 𝐹2 + 2𝒆2 𝐸𝑨𝑨

(𝑈) = λ𝑤

3 λ : sca cattering mean free path 𝒐𝟏 density of the medium 𝒆 size of the acceleration region 𝒃 α λ/𝑜0 𝑮𝟏 injected electron spectrum Suppose λ constant

Strong pitch angle scattering due to small scale magnetic fluctuations ➔ diffusive transport of energetic electrons

Free parameters are 𝒐𝟏, 𝒆 and 𝝁 Deduced from X-rays

Kontar et al. (2014)

Diffusive transport of energetic electrons

Musset et al. (2018)

9

Spectral distribution of energetic electrons Spatial distribution of energetic electrons at 25 keV

Data Model Data Model Corona Footpoints

Bes est fit it to

• bo

both di distributions: 𝒐 = 𝟘. 𝟔 × 𝟐𝟏𝟐𝟏 𝒅𝒏−𝟒 𝒆 = 𝟔. 𝟔 × 𝟐𝟏𝟗 𝒅𝒏 𝝁 = 𝟐. 𝟓 × 𝟐𝟏𝟗 𝒅𝒏

The X-ray observations can be globally explained by the diffusive transport model.

Model fit to spectral and spatial distributions

Musset et al. (2018)

10

Energetic electron density above 25 keV (from X-rays) Energetic electron density above 60 keV Val alues of

• f th

the model par arameters: 𝒐 = 𝟘. 𝟔 × 𝟐𝟏𝟐𝟏 𝒅𝒏−𝟒 𝒆 = 𝟔. 𝟔 × 𝟐𝟏𝟗 𝒅𝒏 𝝁 = 𝟐. 𝟓 × 𝟐𝟏𝟗 𝒅𝒏 𝝁 = 𝟐 × 𝟐𝟏𝟖 𝒅𝒏

To explain both X-ray and radio emissions, need energy-dependent scattering mean free path

Data Model Data Model

Can we also explain radio emissions?

Musset et al. (2018)

11

Probing 2 energy domains (with X-ray and radio) ➔ Energy dependence of electron scattering mean free path, decreasing with increasing energy

Scattering mean free path dependence on energy

Musset et al. (2018)

12

Probing 2 energy domains (with X-ray and radio) ➔ Energy dependence of electron scattering mean free path, decreasing with increasing energy Similar dependence in this energy range found by Dröge (2000), Agueda et al (2014) for energetic electrons in the interplanetary medium

Agueda et al. (2014)

Scattering mean free path dependence on energy

Musset et al. (2018)

12

• Pioneering scattering theory: quasi-linear approach (e.g. Jokipii 1966)
• New approaches take into account dissipation range / wave propagation and wave damping…

Dröge (2003)

Explains mean free path energy dependence for electrons in the interplanetary medium (e.g. Dröge 2003) Note that energy (or rigidity) range is similar to range considered in this talk

Why is the mean free path decreasing with increasing energy?

13

Conclusion on diffusive transport in coronal loops

14

X-ray imaging spectroscopy → transport effect on the distribution of energetic electrons, when a coronal X-ray source is detected. Combination of X-ray and radio diagnostics → energy dependence of transport effects.

Conclusion on diffusive transport in coronal loops

On our study case, a diffusive transport due to magnetic field fluctuations can globally explain X-ray and radio observations. Energetic electrons are trapped in the corona. We found a decreasing mean free path with increasing electron energy. This tendency is found for the same energy range for energetic electrons detected in the interplanetary medium and in updated models of particle scattering.

14

X-ray imaging spectroscopy → transport effect on the distribution of energetic electrons, when a coronal X-ray source is detected. Combination of X-ray and radio diagnostics → energy dependence of transport effects.

Conclusion on diffusive transport in coronal loops

On our study case, a diffusive transport due to magnetic field fluctuations can globally explain X-ray and radio observations. Energetic electrons are trapped in the corona. We found a decreasing mean free path with increasing electron energy. This tendency is found for the same energy range for energetic electrons detected in the interplanetary medium and in updated models of particle scattering. This kind of observation = only when a strong X-ray coronal source in observed, introducing an

• bservation bias.

14

X-ray imaging spectroscopy → transport effect on the distribution of energetic electrons, when a coronal X-ray source is detected. Combination of X-ray and radio diagnostics → energy dependence of transport effects.

Conclusion on diffusive transport in coronal loops

On our study case, a diffusive transport due to magnetic field fluctuations can globally explain X-ray and radio observations. Energetic electrons are trapped in the corona. We found a decreasing mean free path with increasing electron energy. This tendency is found for the same energy range for energetic electrons detected in the interplanetary medium and in updated models of particle scattering. This kind of observation = only when a strong X-ray coronal source in observed, introducing an

• bservation bias.

We need an X-ray imager and spectrometer with a better sensitivity and dynamic range to widen the variety of flares and study transport effect in the case of a faint coronal source. ➔ Motivation for focusing optics for solar X-ray diagnostics (see later in this talk)

14

X-ray imaging spectroscopy → transport effect on the distribution of energetic electrons, when a coronal X-ray source is detected. Combination of X-ray and radio diagnostics → energy dependence of transport effects.

• Launched in 2002
• Decommissioned in 2018

RHESSI

Reuven Ramaty High-Energy Solar Spectroscopic Imager Indirect imaging method: Rotating Modulation Collimators

Need better sensitivity and dynamic!

Difficult to observe small flares, faint sources near bright sources…

15

Why do we need better sensitivity?

Study small-scale events (“nanoflares”) can help us to place better constraints on coronal heating models

• Small flares
• Active regions

nano micro large

Observe flare acceleration region And other faint sources (e.g. energetic electrons in jets, from escaping electron beams in the corona…)

16

• On Solar Orbiter (launch in 2020)

STIX

Spectrometer/Telescope for Imaging X-rays Indirect imaging method: Measure of visibilities with moiré patterns Small and stable background ➔ improvement on sensibility Dynamic range similar to RHESSI

17

• FOXSI sounding rocket: 2012, 2014, 2018
• FOXSI SMEX (NASA small explorer):

phase A (if selected, launch 2022)

FOXSI

Focusing Optics X-ray Solar Imager Direct imaging method: Incidence grazing optics

18

The FOXSI rocket program

NASA sounding rockets:

An opportunity to demonstrate capabilities of new technologies for space applications Benefits:

• Lower cost
• Increases “technology readiness level” of experiment
• Opportunities for student involvement in hardware!

FOXSI launches:

• November 2 2012
• December 11 2014
• September 7 2018

19

The Focusing Optics X-ray Solar Imager

OPTICS 7-shell telescope Replicated Ni optics from NASA/MSFC DETECTORS ▪ 7 double-sided Si or CdTe strip detectors from ISAS ▪ Si: 75 um pitch, 500 um thick ▪ CdTe: 60 um pitch, 500 um thick ▪ Read out by low-power, low-noise ASICs ▪ 7 modules: nested sets of 7 or 10 ▪ FWHM ~5” 20

FOXSI-1

RHESSI

4-15 keV

FOXSI-1

4-15 keV

Launched on Nov 2 2012 FOXSI-1 produced the first focused image of the Sun in HXRs!

• RHESSI image map shows artifacts across entire image.
• FOXSI (direct imaging!) is free of these artifacts.

21

FOXSI-2

Upgrades from first flight:

• Optics modules with greater effective area
• Detectors with higher efficiency at higher x-rays energies
• Solar pointing system aligned with experiment optics

Observations during second flight:

• Two solar microflares
• Several active regions

FOXSI allows us to detect and perform spectral analysis on microflares which are an order of magnitude fainter than flares observed by previous x-ray instruments!

GOES (blue) - Hannah et al. 2008 RHESSI (red) - Hannah et al. 2008

FOXSI-2 (purple) – 1st microflare (5 pointings)

Launched on Dec 11 2014 22

FOXSI-2 spectral analysis of flares

FOXSI-2 first microflare: Estimated GOES class: A0.5 Spectroscopy: isothermal model

Flare #1 Flare #2

W/m2

GOES RHESSI

Vievering et al. (in prep)

23

FOXSI-2 spectral analysis of flares

(35, -230) (30, -230) (27, -231) (18, -232) (38, -234) (29, -232)

Low E: High E:

(33, -239) (25, -233) (31, -239) (25, -228)

Imaging spectroscopy: Centroids at higher energy are located east of the low-energy, suggesting high temperature plasma (energy release) FOXSI-2 first microflare: Estimated GOES class: A0.5 Spectroscopy: isothermal model

Flare #1 Flare #2

W/m2

GOES RHESSI

Vievering et al. (in prep)

23

FOXSI-2 spectral analysis of flares

Can a non-detectable non-thermal population explain the thermal emission? Volume estimation from the FOXSI deconvolved image Temperature and emission measure from the isothermal fit

𝐹𝑢ℎ ~ 𝟐𝟏𝟑𝟖 − 𝟐𝟏𝟑𝟗 𝐟𝐬𝐡

Vievering et al. (in prep)

24

FOXSI-2 spectral analysis of flares

Can a non-detectable non-thermal population explain the thermal emission? Volume estimation from the FOXSI deconvolved image Temperature and emission measure from the isothermal fit

𝐹𝑢ℎ ~ 𝟐𝟏𝟑𝟖 − 𝟐𝟏𝟑𝟗 𝐟𝐬𝐡

For illustrative purposes

Nonthermal

Hypothesis:

• The nonthermal energy is equal to the thermal energy
• The nonthermal spectrum is a thick target spectrum

Values of the low energy cutoff between 3 and 5 keV And values of the slope between -7 and -9 ➔ Dominant thermal spectrum AND enough non-thermal energy to account for the thermal energy

Low energy cutoff Spectrum slope

Vievering et al. (in prep)

24

FOXSI-2 DEM analysis of microflares

Athiray et al. (in prep)

Combined DEM analysis using SDO/AIA, Hinode/XRT and FOXSI ➔ Good constrain on high temperature plasma

25

FOXSI-2 DEM analysis of microflares

Microflares have excess emission above 5 MK Thermal energy estimates:

Multithermal Isothermal Microflare 1 3.3−1.3

+1.7 × 1028 ergs

0.6 × 1028 ergs Microflare 2 4.4−1.1

+3.4 × 1028 ergs

0.9 × 1028 ergs

Multithermal DEM provides estimates of the thermal energy approx. 5 times higher than the isothermal approximation ➔ Small scale energy releases are important to consider for coronal heating

Athiray et al. (in prep)

26

FOXSI-3 overview

FOXSI-3 Science

Characterize small-scale energy release and impulsive heating in the solar corona at high energy

Need to:

• Increase the number of active regions
• bserved by FOXSI
• Increase precision at higher

temperatures

• Increase sensitivity to set a more

stringent upper limit on the quiet Sun

27

FOXSI-3 overview

More 10-shell optics modules ➔ Higher effective area Collimators ➔ Reduce ghost-ray background New CdTe detectors ➔ Higher efficiency > 10 keV PhoEnIX SXR detector ➔ Increase energy coverage

FOXSI-3 upgrades FOXSI-3 Science

Characterize small-scale energy release and impulsive heating in the solar corona at high energy

Need to:

• Increase the number of active regions
• bserved by FOXSI
• Increase precision at higher

temperatures

• Increase sensitivity to set a more

stringent upper limit on the quiet Sun

27

28

FOXSI-3 launch

FOXSI-3 launch

2018 September 7

29

FOXSI-3 targets

1 Active region 134 seconds 2 North Pole 27 seconds 3 East limb 147 seconds 4 Active region 63 seconds 6’11’’ of observation 30

1. Aged active region – 134 s 2. North pole – 27 s 3. Eastern limb – 147 s 4. Aged active region – 63 s ~ 6 minutes observation time

Aged Active Region (134s)

HXRs

North Pole (27s)

HXRs

Eastern Limb (147s)

HXRs FOXSI data overlaid on AIA Fe XVIII images Images show combined data from 3 Si detectors (FOV shown as red squares)

No clear evidence of localized emission observed in HXRs during flight, but more detailed analysis is underway to search for any excess emission above background.

FOXSI-3 HXR observations (preliminary)

31

PhoEnIX Full Sun Mosaic 0.5-5 keV First soft X-ray images of the Sun produced via single photon counting! Spectroscopic images → Time series and energy spectra available for every pixel in the image

PhoEnIX Observations (preliminary)

32

FOXSI-3 coordinated observations: preliminary images

NuSTARcontours (>2 keV) over AIA 94 A image Hinode/XRT image IRIS scan, 1330 A

Goals for FOXSI-3 Analysis

• Determine the temperature structure

and heating mechanisms for an aged active region

• Measure the temperature structure of

bright points observed in EUV and X-rays

• Place limits on non-thermal emission

from the quiet Sun With this unique data set, we intend to:

33

Conclusions

RHESSI has been a successful source of observations for the field of high energy solar physics and the data will continue to provide useful inputs to understand particle acceleration and transport in the solar corona.

34

Conclusions

RHESSI has been a successful source of observations for the field of high energy solar physics and the data will continue to provide useful inputs to understand particle acceleration and transport in the solar corona. However, the next big steps will be achieved with high-sensitivity X-ray instruments. STIX will provide full-disk observations with a better sensitivity and time cadence than RHESSI, with observation windows of several days.

34

Conclusions

RHESSI has been a successful source of observations for the field of high energy solar physics and the data will continue to provide useful inputs to understand particle acceleration and transport in the solar corona. However, the next big steps will be achieved with high-sensitivity X-ray instruments. STIX will provide full-disk observations with a better sensitivity and time cadence than RHESSI, with observation windows of several days. However, only a direct imaging telescope will provide the dynamic range needed to see faint X-ray sources in the presence of bright X-ray footpoints. The FOXSI sounding rocket has demonstrated the technology to observe solar HXR.

34

Conclusions

RHESSI has been a successful source of observations for the field of high energy solar physics and the data will continue to provide useful inputs to understand particle acceleration and transport in the solar corona. However, the next big steps will be achieved with high-sensitivity X-ray instruments. STIX will provide full-disk observations with a better sensitivity and time cadence than RHESSI, with observation windows of several days. However, only a direct imaging telescope will provide the dynamic range needed to see faint X-ray sources in the presence of bright X-ray footpoints. The FOXSI sounding rocket has demonstrated the technology to observe solar HXR. FOXSI SMEX was in the competitive phase A (development phase) in 2018, selection is due “now”. If selected, FOXSI will be launched in 2022 !

34