1. Yael Raveh (Selected Topics in Astrophysics’ Seminar, Spring 2014) X-ray Spectroscopy of Cooling Clusters 02-July-14 1 X-ray Spectroscopy of Cooling Clusters

2. Outline 02-July-14 X-ray Spectroscopy of Cooling Clusters 2 Introduction : Physics of the ICM An ionized plasma – coronal approximation Radiation emission processes – MHD equations What are cooling flows? Means of observation; X-ray spectra of cooling clusters The cooling flow problem Some of current alternative models

3. Clusters of Galaxies 02-July-14 X-ray Spectroscopy of Cooling Clusters 3 Total masses of 10 14 -10 15 ʘ 16% gas, 13% in ICM. 84% dark matter Gas densities : 10 -1 -10 -3 cm -3 X-ray luminosities range : 10 43 -10 46 erg s -1

4. Physics of the ICM 02-July-14 X-ray Spectroscopy of Cooling Clusters 4 The ICM is plasma that is nearly fully ionized → Well-described by MHD theory Electromagnetic radiation emission, mostly X-ray Well-described by the coronal approximation - allows us to observe and study the ICM - X-ray emission tend to cool the plasma-’Cooling Flows’ ייצור קרינת רנטגן ב ICM (  פלזמה מיוננת ) מתואר ע " י הקירוב הקורונלי.

5. Physics of the ICM 02-July-14 X-ray Spectroscopy of Cooling Clusters 5 Once the ionization balance is determined, the X-ray spectrum can be calculated by considering radiation processes The charge state abundance (elemental abundance times fraction ionic abundance) of various ions as a function of temperature. The top panel shows helium-like and hydrogen-like charge states of various low Z atoms. The bottom panel shows iron ions having the outer electron in the K, L and M shell. The bottom panel indicates how the measurement of various ions in the iron series is a sensitive probe of whether plasma at a given temperature exists. Figure uses data from [9] Ionization balance ( לוח ?) Steady state בתמונה – ionization balance vs. electron temperature

6. X-ray Cooling Function 02-July-14 X-ray Spectroscopy of Cooling Clusters 6 An integration of the emission from all processes weighted by the energy of the photons d α /dE is the energy dependent line power (or continuum power) Cooling time (gas enthalpy divided by the energy lost per unit volume of the plasma): In the cores the cooling time approaches cooling times below 5×10 8 yr The relative distribution of plasma at a set of temperatures is often expressed in terms of an emission measure distribution Another convenient way to express the distribution of plasma temperatures – differential luminosity יחידות cooling function : [ergs cm 3 s -1 ] If the gas was undisturbed, it would have a chance to cool several times. Note that as gas cools at constant pressure then the rise in density as the temperature drops means that t cool becomes shorter and shorter

7. The radiative Cooling Function 02-July-14 X-ray Spectroscopy of Cooling Clusters 7 The X-ray region from 10 6 to 10 8 K is dominated by bremsstrahlung The X-ray region from 10 6 to 10 8 K is dominated by bremsstrahlung at high temperatures as well as signi fi cant contribution from line emission at lower temperatures. Below 10 6 K in the UV temperature range, the cooling function rises signi fi cantly solar abundances one-third solar abundances pure hydrogen and helium

8. Magneto-hydrodynamics eqs. 02-July-14 X-ray Spectroscopy of Cooling Clusters 8 A description of the plasma of the ICM on large scales 1. Mass conservation eq. : 2. Navier-Stoker eq. – momentum conservation : 3.Induction eq. – evolution of the magnetic field : 4. The energy eq. : 5. Gravitational field :

9. Cooling Flows 02-July-14 X-ray Spectroscopy of Cooling Clusters 9 The energy loss is directly due to the observed X-ray emission Cooling time : high redshift ~ 10Gyr low redshift ~ 3 Gyr A simplification of the MHD eqs. : Assumptions : 1 th TR law → - X-ray luminosity is heat loss Entropy : - No heating‘Standard isobaric - Steady statecooling flow model’ : - Extra : spherical geometry, atomic physics determines L and T, Locally maxwellian, no absorption, metal distribution, Exact prediction for mdot depends on grav. potential In order to understand what the ‘cooling- fl ow problem’ is, why heating is required, how a ‘residual fl ow’ might operate and what happens when heating is not operating, we now brie fl y examine cooling fl ows If the gravitational fi eld is relatively smooth as can be expected in dark matter haloes, then the pressure will remain nearly constant in small regions that will begin cooling  dp/dT term in 1th equation is small

10. Mass flow rate determination 02-July-14 X-ray Spectroscopy of Cooling Clusters 10 +thermal instability and multiphase flow

11. Cooling Flows 02-July-14 X-ray Spectroscopy of Cooling Clusters 11

12. X-ray instrumentation and observational techniques 02-July-14 X-ray Spectroscopy of Cooling Clusters 12 Important instrument charactaristics : Spectral resolution Effective area and exposure time Field of view Spectral bandpass Angular resotion Chandra X-ray Observatory X-ray XMM-Newton FPCS : First HR observations - FPCS The RGSs on XMM-Newton Analysis techniques : ראינו בהרצאות קודמות ( אוריה ) The data analysis is usually quite complex compared to the techniques that can be applied to data from unresolved sources. In addition, the instrument response functions are actually quite complex, and therefore the application of these functions is usually problematic and results often are somewhat model dependent RGS - re fl ection grating spectrometer

13. X-ray Spectra of Cooling Clusters 02-July-14 X-ray Spectroscopy of Cooling Clusters 13 X-ray Spectrum dominated by line emission and Bremmstrahlung from collisionally ionized plasma Plasma out of LTE optically thin גרף : The X-ray spectrum of the central region of the Virgo cluster of galaxies obtained with the ASCA observatory highest quality spectra obtained prior to the launch of the instruments on XMM-Newton השתמשו ונוסו מודלים ספקטרליים רבים ע " מ לייצג פליטת קרני רנטגן מ cooling flows. the relatively low spectral resolution made it dif fi cult to distinguish between models הבליטות בגרף נובעות מרזולוציה נמוכה – לא ניתן להבחין בין קווי פליטה קרובים (In particular, the bump near 1 keV is due to a forest of Fe L shell transitions, which are quite sensitive to the temperature of the plasma)  X-ray Spectra (prior to 2000) LTE = Local Thermodynamic Equilibrium Collisional ionizations balanced by recombinations Line emission dominated by collisional excitations+cascades, Radiative recombination, and dielectronic recombination Same model as stellar coronae At densities and temperatures (in core), t recombination = 10 6 years (for Fe XVII at 1 keV) t cool = (5/2 n k T)/(n 2  ) = 10 8 to 10 9 years t formation = 5 10 9 years

14. X-ray Spectra of Cooling Clusters 02-July-14 X-ray Spectroscopy of Cooling Clusters 14 Long-standing prediction that cores of clusters should cool by emitting X-rays in less than a Gyr Temperature DropsDensity rises and t cool is short

15. The Cooling Flow Model vs. Observations 02-July-14 X-ray Spectroscopy of Cooling Clusters 15

16. Failure of The Model 02-July-14 X-ray Spectroscopy of Cooling Clusters 16 Hot Clusters (4-10 keV)

17. Failure of The Model 02-July-14 X-ray Spectroscopy of Cooling Clusters 17 Warm Clusters (2-4 keV)

18. Failure of The Model 02-July-14 X-ray Spectroscopy of Cooling Clusters 18 Cold Clusters (1 to 2 keV)

19. 02-July-14 X-ray Spectroscopy of Cooling Clusters 19 Differential Luminosity vs. Temperature Differential Luminosity vs. Fractional Temperature

20. Differential Luminosity ~ T   ~ 1 to 2 02-July-14 X-ray Spectroscopy of Cooling Clusters 20 Some of the observational results : Model fails at a fraction of T max rather than fixed T~1keV Model fails in shape as well as normalization; Tilted toward higher temperatures Some scatter in both slope and normalization Unclear if relation continues to low temperature for all clusters or not

21. The Cooling Flow Problem 02-July-14 X-ray Spectroscopy of Cooling Clusters 21 The ‘mass sink’ problem – in some objects, the total star formation rate (~Gyr) and gas mass is 1-2 orders of magnitude below the enormous implied mass deposition, A simple cooling fl ow spectrum* is a poor fi t to the data in the soft X-ray band * High spectral resolution XMM/RGS data / Chandra spectra

22. Some of Current Models 02-July-14 X-ray Spectroscopy of Cooling Clusters 22 Heat Conduction conductivity appropriate for an unmagnetized gas : κ S ∝ T 5/2 Effective conductivity in clusters with a complex magnetic fi eld topology ??

23. 02-July-14 X-ray Spectroscopy of Cooling Clusters 23 Heating by a Central Radio Source – radio bubbles radii ~ 1–15 kpc Energy : E min =PV – E max exceeds 10 61 erg Lt expansion =PV ∝ PR 3 L – power of jet → V bubble ∝ t -2/3 High resolution X-ray image - the Perseus cluster of galaxies. The bright source in the center is the AGNof NGC 1275. The two adjacent holes in the X-ray emission as well as the mushroom shaped depression in the upper right are believed to be cavities in the ICM that have been excavated by cosmic rays expelled from the active galactic nucleus. Numerical simulations of Radio Bubbles in ICM : Several assume that the surrounding gas is isothermal. Many produce bubbles which are very unstable, which is most unlike the observed bubbles. Indeed most simulated bubbles do not look like the observed ones. A catastrophic cooling fl ow is only being postponed for a while. Although this explanation can work temporarily it cannot provide a comprehensive solution. Heating by a mixture of weak shocks and viscous damping of sound waves - Transport of the energy by sound waves and dissipating it provides a fairly gentle and distributed source of heat. The bubbles in a viscous medium also better resemble the observed ones

24. Summery 02-July-14 X-ray Spectroscopy of Cooling Clusters 24 Despite very simple theoretical arguments, Cooling flow model fails to reproduce X-ray spectrum; Several strong observational constraints Much more theoretical and observational work needed for fine- tuning challenges There still remains the possibility that some process not yet foreseen, however the central radio source is so common and energetic that it must at least be part of the solution.