Lab. for High Energy Astrophysics
Goddard Space Flight Center
Astronomy Department
University of Maryland
with thanks to :
A brief backward look.
What we can measure with current technology.
What observational and physical issues to watch out for in turning the observed spectrum into elemental abundances.
The new results from ASCA including :
Some theoretical implications and future prospects.
Non-X-ray spectroscopic methods of measuring abundances in the ICM.
Issues relating to the energetics of the ICM.
To quote Renzini etal. (1992) :
``...an uncommon variety of astrophysical issues converges on the quest for iron (and I would add, other elements) in clusters, from the formation and evolution of elliptical galaxies and clusters of galaxies to supernova physics and astrophysics and from nucleosynthesis and chemical evolution in the halo and disk of our Galaxy to the evolution of gas flows and X-ray properties of elliptical galaxies, with topics such as star formation, dark matter, AGN activity, galaxy interactions with the environment, and many others coming into play.''
Clusters of galaxies comprise three components, the individual galaxies, a hot plasma occupying the space between the galaxies, and something else providing most of the mass.
The hot plasma has temperatures of 10^6 - 10^7 K and densities of 10^-2 - 10^-5 electrons/cm^-3. The plasma is collisional with X-ray emission from optically-thin, thermal bremsstrahlung and emission lines. X-ray luminosities range from 10^43 to a few times 10^45 ergs/sec.
The mass in plasma exceeds that in stars, in rich clusters by a factor of five. Most of this plasma must be primordial, having been part of the cluster when it first collapsed or having been accreted subsequently.
The plasma is static except in the cores of clusters where cooling flows are seen and in clusters that are undergoing mergers or accreting sub-clusters.
Between the energies of 0.5 and 10 keV and for a plasma at the temperatures characteristic of clusters of galaxies the main emission lines observed are from transitions to the n = 1 state (K-shell) for He-like and H-like ions.
The exception is Fe which has a large number of emission lines to the n = 2 state (L-shell) for ion stages from Fe XVI to Fe XXV. These cover the energy range 0.7 -- 1.5 keV.
The elements with strong enough lines to be observed with present technology are O, Ne, Mg, Si, S, Ar, Ca, Fe, and Ni.
The Fe K-shell lines have large equivalent widths and are in an isolated part of the spectrum so Fe is observationally easy and reliable to measure.
Si, S, Ar, and Ca are all in isolated parts of the spectrum but have smaller equivalent widths so these are reliable but harder to measure.
O, Mg, and Ne are in the energy range dominated by the Fe L-shell lines. With current X-ray spectrometer resolutions these elements cannot be measured independently of the iron lines. Thus, abundance measurements for these elements are more susceptible to systematic problems.
Ni measurements should be reliable but the Ni K lines have smaller equivalent widths than Fe K and current telescopes and detectors have low efficiencies in this energy range.
Ionization equilibrium. To accurately measure abundances we need to be confident that the plasma is not over- or under-ionized for its continuum temperature. Both ionization and recombination timescales are < 10^9 yrs for the temperatures and densities of the ICM.
Transition strengths. Most of the important lines are K-shell transitions of He-like or H-like ions. These are well understood theoretically and extensively observed in the Solar corona. The Fe L-shell lines between 0.7 and 1.5 keV are less secure and may have systematic problems.
Radiation transfer effects. The continuum emission is optically-thin everywhere in the cluster. There may be significant effects due to resonant absorption and electron scattering in the very central parts of the cluster (Wise & Sarazin 1993).
It is important that the ICM be isothermal or that the temperature structure be well determined.
Current spectral resolutions mean that many lines are blended together requiring accurate plasma emission codes. In particular, Mg and Ne are in amongst the Fe L-shell lines.
Detectors have a number of spectral features associated with elements in the X-ray path or the reflection surface of the optics. The ASCA telescopes produce Au M edge features while the CCDs show both O and Si edges from the gate structure.
The heavy elements in the ICM are produced by SN in the elliptical galaxies in the cluster. (M.Arnaud etal. 1992)
The observed abundance ratios imply that most of the heavy elements come from SNII although S, Ar, and Ca abundances are all low.
Assuming the ICM is homogeneous then :
Approximately 150 clusters have been observed by ASCA so far. All of these will yield at least an iron abundance. Many will enable measurements of other elements. All this data will end up in the public archive, accessible electronically.
In the longer term Astro-E (launch in 2000) will provide 10 eV resolution spectroscopy. Among other advantages this will mitigate the problems associated with the Fe L lines.
Abundance measurements of other, less common, elements may become possible with resolutions of 2 eV (NGXO).
Isotopic abundances are likely to be impossible because of thermal broadening of lines.
We are at the beginning of a new era for the measurement of abundances in the ICM.
The first examples of the new results that this will bring are :
Clusters show varied behaviour in their spatial distribution of heavy elements. Some have central peaks, others do not.
The iron abundance in the ICM has been measured out to z~0.5 with no evidence of variation with redshift.
Reliable abundances of alpha-burning elements have been measured in selected clusters. The results demonstrate that SNII were an important source of the heavy elements now seen in the ICM.