X-RAY SPECTRAL SIGNATURES OF PHOTOIONIZED PLASMAS
Liedahl, et al.
ApJ, 350:L37-L40, 1990 February 20

I.	INTRODUCTION

BACKGROUND

-collisional equilibrium often assumed in plasma emission codes
-this assumption makes the codes inapplicable to most situations
 in astrophysics (namely, x-ray photoionized nebulae)

-x-ray plasmas exist in both hot diffuse gas and cool dense 
 accretion flows in X-ray binaries
-spectroscopy in soft x-rays probes extreme environments of 
 x-ray plasmas

COLLISION-DOMINATED X-RAY PLASMA

-heating by mechanical processes
-ionization by electron-ion collisions
-excited atomic levels populated by electron impact
-abundance peaks of soft x-ray emitting ions occur at T > 10K
-ionization balanced by dielectronic recombination
	http://www.astro.umd.edu/~chris/thesis/node148.html

NOTE:  dielectronic recombination is an alternative to 
radiative recombination (in which excess electron energy is
radiated away, either as continuum or line photons) in which
excess electron energy is used to excite an ionic electron.
Following this process the ion is doubly-excited, which is 
where this recombination process gets its name: dielectronic.

-optically thin plasmas are coronal 
	(used for most emission codes)

PHOTOIONIZATION-DOMINATED X-RAY PLASMA (XPN)

-strong source of ionizing radiation effects state of gas
-generally overionized relative to local electron temperature
-collisional excitations out of the ground state are rare since
 electrons don't have enough energy
-main sources of excitation are radiative recombination or cascade
 following recombination


BAD CODES AND MODELS

-because of differences between collisional plasmas and 
 photoionized plasmas, the coronal plasma emission codes are not
 applicable for analyzing the spectra of XPN sources
-lack of suitable spectra models for XPN has hampered analysis of 
 measured spectra

GOAL

-focus attention on the different line-excitation mechanisms of 
 coronal and photoionized plasmas, and see whether these differences
 are evident in the spectra
-to identify differences, they compare the XPN spectrum with the 
 spectrum calculated for the same ions under coronal equilibrium

RESULTS

-they show that the 3d-2p transitions (which are strong in the Fe
 L-shell spectra from coronal plasmas) should not be present in 
 spectra from phoionized plasmas
-in the spectra from photoionized plasmas, the strongest lines 
 should be from the 3s-2p transition
-qualitative difference between the two can be used a method of 
 identification and differentiation

II.	ATOMIC PHYSICS CALCULATIONS

WHY IRON?

-emission lines from several ionization stages of Fe are expected 
 to be prominent in x-ray spectra from cosmic sources
-in fact, they these lines can be among the brightest in the spectra
 from these sources

CALCULATION

-calculated model spectra for the Fe XVI-XIX ionization stages for 
 two different environments:
	
	(1)	hot coronal plasma in collisional equilibrium;

	(2)	an XPN.

-Assumptions:

		CORONAL			XPN
 Temp (k*T_e)	500 eV			10 eV
 Density	10^11 cm^-3		10^11 cm^-3

-the rate equations are solved in steady state for the excited 
 level populations
-observations suggest that higher densities are not realistic for
 these environments, although the are completely accessible since 
 the calculations are collisionally radiative
-atomic structure, radiative decay rates, and rates for electron
 impact excitation calculated using HULLAC atomic physics 
 package
-larger atomic models used in the XPN case than in the coronal 
 case because cascades are important following radiative 
 recombination

NOTE:  Following the Osterbrok formulation, radiative recombination
is a cooling process, since it removes an electron from the 
"electron sea", thereby removing energy from the plasma, of the 
magnitude k_b*T.

-the n = 5 and n = 4 levels are not included in the calculation of
 the collision rates because they cannot be reached collisionally 
 from the ground state given the low temperatures of the XPN
 evironment
-dielectronic recombination rates are calculated to measure the 
 importance of this line excitation process in the measured spectra 
 of XPN's
-they conclude that DR processes do not effect the X-ray line 
 spectrum
-radiative recombination cross sections taken from photoionization
 cross sections

III.	RESULTS

-FIG 1: coronal plasma
-FIG 2: XPN
-both only look at line emission from Fe for the ionization states
 XVII - XX and more to the point, only iron and no other elements
-Once again, they conclude

	coronal plasma --> 3d-2p transitions
	XPN	       --> 3s-2p transitions

-some lines in both are well-known components of solar x-ray spectra
-so, XPN are characterized by their lack of 3d-2p transitions
-the explanation for the lack of these lines is the low collision 
 rate to 3d states
-nearly all flux into the n = 3 shell goes to the 3s state, whereas 
 for a coronal plasma much of the flux into n = 3 shell goes to the 
 3d state through collisions (electron impact excitation)

IV.	DISCUSSION

-Ratio of (3s-2p)/(3d-2p) can be used to determine the principal 
 mode of line excitation
-Ratios of 3s lines from different ionic species can determine an
 ionization balance
-XRB's cannot be spatially resolved so that their composite spectrum
 may result from independent coronal and photoionized regions
-thus, being able to disentangle the contributions of the composite
 parts is vital to understanding the geometry and physical processes
 of accretion flows    

QUESTIONS

What do the authors mean when they say "Furthermore, at densities 
likely to be encountered in astrophysical plasmas, these states 
will not be collisionally mixed due to the ready availability of 
"allowed" radiative decay channels"?

In general, what is meant by a channel?