1 / 33

From Quasi-Periodic Oscillation models to a microquasar classification

From Quasi-Periodic Oscillation models to a microquasar classification. P. Varniere (APC) et al. Summary. Introduction  Basic microquasar observations  What should a model for Low Frequency / High Frequency Quasi Periodic Oscillation explain? AEI as a model for LFQPO

jin
Télécharger la présentation

From Quasi-Periodic Oscillation models to a microquasar classification

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. From Quasi-Periodic Oscillation modelsto a microquasar classification P. Varniere (APC) et al.

  2. Summary Introduction Basic microquasar observations What should a model for Low Frequency / High Frequency Quasi Periodic Oscillation explain? AEI as a model for LFQPO Basics characteristics of the AEI AEI as a LFQPO model Magnetic Flood Scenario and Disk Instabilities The Magnetic Flood Scenario what is at the origin of the A,B, C states in the Flood Scenario RWI at the origin of the state B? Disk Instabilities and states classification three instabilities but four states Another look at this classification: variability based classification a quick look at the LFQPO-types Conclusion

  3. Microquasars observations a microquasar is a binary system hosting a black-hole of a few tens of solar mass being “fed” by an accretion disk and emitting relativistic jets. Microquasars appear to transit between several states, here we are mainly interested in the ones displaying timing variability known as Quasi-Periodic Oscillations (QPOs) The Steep Power-Law state involves disk and power-law emission are both present Low Frequency Quasi Periodic Oscillation(LFQPO) andHigh Frequency Quasi Periodic Oscillation (HFQPO) The low-hard state involves mostly power-law spectrum emission of steady jet and Low Frequency Quasi Periodic Oscillation Steep Power-Law Thermal state Low-Hard state large variety of observations (X-rays, Gamma rays, Infra-Red, radio,...) and technics (timing, lags) ➜ many constraints on theories

  4. Looking for a LFQPO model: what we need to explain ✵variable frequency between 1-30 Hertz, with a fast change ✵ coherence⇒ QPO can't be due to blobs ⇒ global organized motion of the gas = normal modes ✵ rms amplitude as much as 30% ⇒ QPO can't happen by random process (blobs) ⇒ must be due to strong disk (+ corona ?) instabilities ✵ correlation of the LFQPO frequency with the soft flux ✵QPO associated with a state where the power-law is strong ✵ the existence of several “types” of LFQPO with distinct characteristics (A, B and C) ✵ lag sometimes changing sign, sub-harmonic structure...

  5. Looking for a HFQPO model: what we need to explain ✵small variation in the observed frequency (maximun change in frequency of about +/- 15%) ✵ HFQPOs appear alone or in “pairs” (with related frequencies, most of the time in a 2:3 ratio, sometimes 1) ⇒ need of one mechanism that can select several linked frequencies depending on the disk condition ✵ HFQPOs can occur in the absence of LFQPOs ✵ when they co-exist we have “unusual” LFQPO (type A and B) ✵ HFQPOs rms amplitudes are much lower than LFQPO and seem anti-correlated with the LFQPO rms ⇒ the HFQPO models need to be coherent with a LFQPO model as they have to co-exist in the disk while being independent

  6. Summary Introduction Basic microquasar observations What should a model for Low Frequency / High Frequency Quasi Periodic Oscillation explain? AEI as a model for LFQPO Basics characteristics of the AEI AEI as a LFQPO model Magnetic Flood Scenario and Disk Instabilities The Magnetic Flood Scenario what is at the origin of the A,B, C states in the Flood Scenario RWI at the origin of the state B? Disk Instabilities and states classification three instabilities but four states Another look at this classification: variability based classification a quick look at the LFQPO-types Conclusion

  7. Basics of the AEI a disk threaded by a vertical magnetic field ~ equipartition  large scale instability a spiral wave ~ galactic spirals but driven by magnetic stresses rather than self-gravity same structure of large scale normal modes (= standing patterns) as in galaxies the frequency of the m=1 mode of the spiral is a few tenths of the rotation frequency at the inner edge of the disk which is coherent with the observed range we need to compare in more detail with observation

  8. Frequency-Radius correlation if we look at the evolution of the color radius and of the LFQPO frequency a correlation seems to exists in objects like GRS 1915+105 and XTE J1550-564 ➜ we decided to look at the AEI prediction 1 obs per points At the same time we were working on the correlation, a paper by Sobzack et al came out showing the distinct behavior of GRO J1655-40 compared to XTE J1550-564 concerning the frequency-radius correlation ➜ we decided to compare those observations with the AEI prediction 16sec per points 4sec per points 8 the frequency evolution of the AEI seems coherent with the observed one... what about its rms amplitude?

  9. The AEI as a model for the LFQPO ✵frequency between 1-30 Hertz ➥frequency of the spiral wave a few tenths of the rotation frequency at the inner edge of the disk ✵stability in time ➥ large scale structure as in galaxies (quasi-standing spiral structure) ✵ rms amplitude as much as 30% ➥ we observed in the simulation as much as 10% in an oversimplified model, we are working on a better model ✵ correlation with the soft flux ➥ comparison with observation in Varniere et al A&A 2002 and Mikles et al. ApJ 2009 ✵ QPO associated with a state with a strong power-law ➥ accretion energy not deposited locally (not heating the disk) but ejection through Alfven Waves emission ✵ lag sometimes changing sign, sub-harmonic structure ➥ possibility of geometrical effects from the jet

  10. The Magnetic Flood Scenario the AEI is a candidate to explain the LFQPO and can also be linked with ejection. next step : assume that AEI ⇔ QPO … then try to understand .... ➜“Magnetic Floods” scenario : the cycles are determined by gradual accumulation and sudden destruction of magnetic flux we looked at the cycles of GRS 1915+105 because: - we had already worked on that source before - multi-wavelength observations were available - it has known (and highly repeatable) cycles between state C, A and B (Belloni et al 2000)

  11. The Magnetic Flood Scenario The cycle begins in the “high/soft -thermal state” during which magnetic field is accumulated in a turbulent disk (possibly driven by the MRI) ➝ B becomes of the order of equipartition ➜ the disk becomes unstable to the AEI, LFQPO appears. it is in the “Low-Hard state” ➜ at the intermediate peak we have reconnection of the magnetic field, causing an ejection and the plasma β to increase ending the AEI and the LFQPO causing the disk to go back to a softer state ➜ we linked instability occurring in the disk with observational states

  12. What is at the origin of the three states? All the observations from GRS 1915+105 can be classified in 12 classes made of three states labeled A, B and C as shown on the color-color diagram. From the Magnetic Flood Scenario we have: state A is dominated by the disk, from the magnetic flood scenario we associate it with the MRI, namely a turbulent disk. state C is dominated by the corona and a jet is present. The Power spectrum presents a LFQPO. We proposed a disk dominated by AEI to explain those features.

  13. RWI at the origin of the state B? state Bhas both a strong disk and a corona, no strong radio emission are observed andHFQPOs are present in the Power spectrum (the 67Hz) but no LFQPOs. In 2006 we proposed the Rossby Wave Instability (RWI) to be at the origin of HFQPOs in black hole systems. This instability requires having an extremum of vortensity which exists in disks with their inner edge close to the last stable orbit. One interesting characteristic of this instability is that, depending on the inner boundary, the m=1 mode is not dominant but it is rather a mix of the m=2 and m=3 modes that dominates. a disk with an extremum of is necessary possibility of emitting Alfven waves in presence of a low density corona above the disk ➜ we need to see if that instability is a good candidate for the HFQPOs in microquasars and therefore a candidate to explain the state B of GRS 1915+105

  14. ✵ small variation in the observed frequency RWI: toward a model for the HFQPO? ➥the existence of the RWI is linked to the position of maximun of vortensity ✵ seems to select pairs of frequencies, in a 3:2 ratio ➥ depending on the inner boundary, the dominant mode of the RWI is the m=2 / m=3, or the m=1 ✵ the HFQPO model need to be coherent with a LFQPO model ➥ the condition for the magnetized RWI are similar to the one of the relativistic AEI ✵ link with the appearance of unusual (type A and B) LFQPO ➥ We proposed an explanation for the different types of LFQPO based on the behavior of the relativistic AEI ✵ we still have several points to work on before being able to compare the RWI-HFQPO model to observations, especially the computation of the observed rms amplitude, the Alfven wave emission, the variation of frequency and 3D MHD simulation... ⇒We are also working on better understanding the constraints from observations. ➜ this instability was the last piece of the puzzle and we now have three disk instabilities which could explain the three states

  15. Summary Introduction Basic microquasar observations What should a model for Low Frequency / High Frequency Quasi Periodic Oscillation explain? AEI as a model for LFQPO Basics characteristics of the AEI AEI as a LFQPO model Magnetic Flood Scenario and Disk Instabilities The Magnetic Flood Scenario what is at the origin of the A,B, C states in the Flood Scenario RWI at the origin of the state B? Disk Instabilities and states classification three instabilities but four states Another look at this classification: variability based classification a quick look at the LFQPO-types Conclusion

  16. The 3 instabilities at the origin of 3 states All the observations from GRS 1915+105 can be classified in 12 classes made of three states labeled A, B and C as shown on the color-color diagram. state Bhas both a strong disk and a corona, very low radio observed andHFQPOs at ~67Hz in the Power spectrum. This state might be explained by a disk dominated by the RWI. state Ais dominated by the disk. From the magnetic flood scenario we associate it with the MRI, namely a turbulent disk. state C is dominated by the corona and a jet is observed. The Power spectrum presents a LFQPO. We proposed a disk dominated by AEI to explain those features. ➜ this can explain the observed 3 states of GRS 1915+105 but not the presence of both HFQPO and LFQPO

  17. 3 instabilities but 4 states validity RWI: inner edge of the disk close to the last stable orbit and an extremun of: We see that there is a domain where both the AEI and the RWI can be active in the disk. The inner edge is close to the last stable orbit and the disk is at the equipartition. In that state we have both the LFQPO and the HFQPOs AEI: equipartition between gas pressure and magnetic field and we get the following four possible states with our three instabilities: - MRI: turbulent disk, similar to the A state, high/soft state, thermal state... - RWI: hot disk, HFQPO, very low radio, similar to B state, other? - AEI: cold disk, LFQPO, jet, BLN, similar to the C state, low-hard state - AEI-RWI: warmer/hot disk, low radio, HF + LFQPOs, steep power-law, VHS ➜ The fourth state exhibit both LF and HF QPOs. It was not present in the original paper by Belloni et al., but it has since been observed and GRS 1915+105 has become the first object to show all 4 states. This AEI-RWI state is often observed in other microquasars but the only RWI state has not been observed yet.

  18. Variability based classification Another way to look at the previously described classification is through the variability Is there a LFQPO? is the AEI active in the disk? Is there a HFQPO? is the RWI active in the disk? This way to look at the classification is model-independent. It is not important what instability is at the origin of the LFQPO and the HFQPO as long as they can operate simultaneously in the disk as required by observation ➜ strong requirement for both LF and HF QPOs model It is now interesting to look at the behaviour of the AEI in this forth state where both the AEI and the RWI are present in the disk ➜ can it explain the different characteristic of the LFQPO?

  19. A quick look at LFQPOs type The most common LFQPO is the C-type, it happens during the low hard state and varys a lot in frequency B A C the other two (A and B) types of LFQPO occur during the Steep-Power Law state and their frequency does not vary much. they are rarer than the C-type and have not been observed in all objects yet. We are working in an extensive study of their properties in all the objects that have exhibited them. This should help to constrain the models ➜ one constraint that seems to arise from observations of XTE J1859+ 226 (Casella et al. 2004), is the need of smooth transitions from C to B and B to A.

  20. A quick look at LFQPOs type When we are in a state where both the RWI and the AEI exist, it is actually the relativistic AEI which is present. The R-AEI has slightly different properties. We did numerical simulation of a disk having both the AEI and the RWI and it appears that the dominant mode is always for same for both of the instabilities. In that picture: C-type is the “regular AEI” B-type is the R-AEi with the mode m=2 dominating A-type same for the m=3 ➜ work in progress but already some interesting points link between the type of LFQPO and the HFQPO mode ➜ both instabilities (AEI and RWI) are in the same disk, the same mode are dominant for both instability. To test this further, we looked at the probable HFQPO of m=1 mode in XTE J1550-564, broad feature at freq 92 (184-276). During that time the LFQPO was “C-modifed” and close in frequency to the other types. the frequency of the A- and B-type has a small range (compared to the c-type) ➜ to be in the AEI-RWI state, the inner edge of the disk needs to be close to the last stable orbit (within 1.3) so the LFQPO cannot change frequency as much as in the C-type case (sub)-harmonic content (A- and B-type sometimes present sub-harmonics or “almost harmonics”) ➜ contrary to the non relativistic case, it is not the m=1 mode of the AEI which always dominates when the disk get close to the last stable orbit. The non-dominating m=1 could be taken for a sub-harmonic. Also, the different modes of the AEI are in close-harmonic relationship, not exact.

  21. Conclusion ℘ The AEI is a possible explanation for several of the characteristics of the LFQPO (frequency, correlations, stability, link with ejection and the different types) ℘ The RWI need to be further developed and tested as a possible model for the HFQPO ➜ we are also working on making a list of observational constraints for both LFQPO and HFQPO model ℘ using a consistent model : control by the gradual accretion of vertical magnetic flux, stored in the BH magnetosphere and destroyed in reconnection events (= ejections) when disk magnetic flux and stored flux are opposite ➜ we proposed a new classification based on which instability dominates the disk: equivalent to a variability based classification ➜ studying the instabilities behaviour in that new state gives a possible explanation for LFQPO types which need to be further tested ℘ the next step is to look at the consequences of the different instabilities on ejection to also link the four state with the radio emission observed. ℘need more complex models than alpha-disks, ADAF, etc. ℘need other control parameters besides M.

  22. The Observed Lag Structure The phase lag structure between the hard and soft X-ray photons observed in GRS 1915+105 and XTE J1550-564 has been said to be « complex » because : the phase of the low-frequency Quasi-Periodic Oscillation (QPO) fundamental Fourier mode changes sign with time (negative to positive to negative) thesign of the odd and even harmonics lag behaves differently Difficult to explain in the framework of the comptonization model

  23. Aim and Idea Aim: how to have a lag that changes signwithout changing the physical process at the origin of both the soft and hard X-ray how to have the coherence between them dropping without changing the physical process that relate them Idea:adding a new effect (absorption) on the top of the low energy band of the QPO modulation (suppose to come from an orbiting structure coherent with the AEI) behavior of Fourier Transform in case of an absorb signal

  24. Lag: a simple derivation a change in any one bands, caused by internal or external phenomena, can create a sign reversal of the lag Ex: using this in the low energy band,one can reproduce the observed behavior of the lags and harmonics If there is an absorption of a small part of the QPO modulation in the low energy band and not in the hard one this simple fact can explain the changing sign of the lag

  25. GRS 1915+105 look likesome absorption in the low energy band? more constraints from observations...

  26. GRS 1915+105: Timing/X-ray QPO freq total flux thermal flux power-law flux phase lag the temporal properties correlate with the different component of the X-ray flux BUT two populations can be distinguished QPO freq > 2 Hz , usual correlation with the total and power-law fluxes QPO freq < 2 Hz , no correlation (similar flux level)... those points are the only ones exhibit positive phase lag and also have high radio flux

  27. GRS 1915+105: timing/radio phase lag QPO freq radio flux mJy, 15.2 GHz ratio of LF power coherence in the radio-loud state the timing behavior is modified at quasi-constant X-ray flux QPO freq < 2 HZ : associated with high radio flux and positive phase lag  those points also have a low coherence between the soft and hard X-ray(meaning the hard component cannot be deduced from the soft one by a linear transformation) either these QPOs arise from a different mechanism (one related to the jet and the other not) orthere is a threshold in radio flux above which new phenomena appear in addition to the QPO mechanism.

  28. Application to microquasars obervation Using the constraint from observations together with the Fourier Transform behavior in case of an energy dependent absorption. What can produced that absorption: ð need to find what may produce the absorbed part of the QPO modulation at low energy, and need to be related to the jet emission. Suppose the base of the jet/corona gets « between » the observer and the structure that create the modulation. This will induce a small modulation on top. In fact any structure that can absorb a part of the low energy modulation and related to the jet may work. We will continue to study this by using numerical simulation and produce synthetic spectra that will be compare with observations. This simple model is able to explain the changing sign of the lag and the drop of coherence...

  29. absorption? toy model...

  30. Basics of the RWI

  31. The RWI as a model for the HFQPO the RWI require an extremum of in the case of a disk approaching its last stable orbit ( ), the epicyclic frequency has a maximum and usually so does evolution of the pattern speed and growth rate as function of the for the lowest modes. ➥ the instability is stronger in presence of magnetic field ➥ the m=1 mode is not the most unstable In presence of a magnetic field the Rossby vortex will twist the footpoint of field line and emit Alfven Wave

  32. the AEI and the RWI hydrodynamic instability that gets stronger in the presence of a magnetic field MHD instability which requires a fully magnetized disk an extremum of a positive gradient of most often the m=1 is not the dominant mode. It is either the m=2 or m=3 mode that dominates most often them=1 (one armed spiral) dominates if the disk has a low density corona energy and angular momentum from the vortex are transfered upward as Alfven waves to the corona power for a jet or a wind differential rotation+ differential vorticity unstable by coupling to a Rossby vortex (~ great red spot of Jupiter) extracts energy and angular momentum from the disk ( accretion) and stores them in the Rossby vortex

More Related