NOTICE: This Legacy journal article was published in Volume 1, May 1992, and has not been updated since publication. Please use the search facility above to find regularly-updated information about this topic elsewhere on the HEASARC site.

Plasma Emissions Code

Stephen A. Drake

HEASARC

Introduction

About 20 years ago, detailed calculations of the ionization equilibria of elements in low-density, collisionally-controlled plasmas with electron temperatures anywhere from 10⁴ to 10⁸ K were performed (e.g., Jordan 1969; Landini and Monsignori-Fossi 1972; Summers 1974). The immediate object of these calculations was the modeling of the outer atmosphere of the Sun: the chromosphere (Te ~ 10⁴ K), transition region (Te ~ 10⁵ K), and corona (Te ~ 10⁶ - 10⁷ K). The atomic processes included in these calculations were collisional ionization, collisional excitation followed by autoionization, radiative recombination, and dielectronic recombination. The collisional ionization and excitation is assumed to be produced by free electrons with a Maxwellian distribution of energies. The effects of an ambient radiation field were neglected and steady-state conditions were assumed to apply.

The ionization codes were subsequently used to calculate the predicted X-ray and UV spectra of the modeled plasmas. Some of the first spectral calculations were done by Landini and Monsignori-Fossi (1970) and Mewe (1972). Because the additional assumption that all spectral lines are optically thin was made, these spectral codes are generally only valid for plasmas with temperatures Te >~ 10⁵ K. Since these original studies, many refinements have appeared in the literature, including improved atomic physics data, the addition of previously neglected atomic processes, and (in some cases) the detailed calculation of satellite lines needed to model high spectral resolution X-ray spectra.

The Mewe (or Mewe and Gronenschild or MG) code has been extensively documented in six papers (Mewe 1972; Mewe 1975; Gronenschild and Mewe 1978; Mewe and Gronenschild 1981; Mewe et al. 1985, 1986). It can be used to generate a model continuum in the spectral range from 1 to 1000 Angstroms (0.012 - 12.4 keV), and to calculate emission line intensities in the somewhat more restricted range of 1 to 300 Angstroms (0.04 - 12.4 keV). Because of its satellite lines calculation, this code has been much applied to the study of high-spectral resolution solar X-ray spectra such as those obtained by the Solar Maximum Mission's X-Ray Polychromator.

The Landini and Monsignori-Fossi (hereafter LMF) code has also been used to model the spectrum of the solar transition region, corona and loops. Discussions of this code can be found in Landini and Monsignori-Fossi (1970, 1985, 1990). In its latest form it can be used to calculate spectra anywhere in the range of 1 to 2000 Angstroms (0.006 - 12.4 keV).

The Raymond (or Raymond and Smith or RS) code has been extensively used in the community to model X-ray and EUV spectra (from 1 to 1200 Angstroms, or 0.01 to 12.4 keV) of a wide range of astrophysical objects. It is described in Raymond and Smith (1977): see also Raymond et al. (1976) and Raymond (1988). An extended version of this code is also available that includes the calculation of forbidden lines and of hydrogen emission lines and can be thus used to produce model spectra that extend into the infrared region.

A number of other independent calculations of the X-ray spectra from thermal, collisionally-dominated stationary plasmas have appeared in the literature in the last 20 years or so: an (incomplete) list of such studies includes Tucker and Koren (1971), Kato (1976), Shapiro and Moore (1976), Stern et al. (1978), Shull (1981), and Gaetz and Salpeter (1983). Other studies have included relaxing some of the implicit assumptions built into most of the previously cited work, by including for example time-dependence or the effects of an external radiation field, but these by their very nature tend to be designed for specific astrophysical situations such as supernovae, and will not be discussed in the present short article.

Rationale for Making Plasma Emission Codes Available at the HEASARC

We have started assembling at the HEASARC some of the above-mentioned thermal plasma emission codes. We have been motivated in this regard by several considerations; difficulties and confusion have periodically arisen when such 'standard' codes have been used in the analysis of X-ray spectra. The considerations are:

(i) To introduce version control

Often an author will make the bald statement "I have modeled the spectrum using a Raymond and Smith thermal plasma and find.....". The parameters that the user infers from such an analysis may or may not be reproducible by others depending on whether they have the same version of the code as he does, since over the decades that these codes have been widely distributed they may have been fairly extensively revised and updated. By introducing a version control protocol, we hope to be able to eliminate such ambiguities.

(ii) To facilitate intercomparisons of different thermal codes

(iii) To provide tools with which to analyze high-resolution x-ray spectra of astrophysical sources

Forthcoming X-ray observatories like the Advanced X-Ray Astrophysics Facility (AXAF) will obtain high-resolution (E/[[Delta]]E ~ 2 x 10³) X-ray spectra of a large number of astrophysical X-ray sources, compared to the small number of objects observed by the equivalent spectrometers on the Einstein and EXOSAT observatories, because of the large increase (~10³) in sensitivity. It is clearly essential to have model spectra produced by plasma emission codes that are of similar high resolution.

Note that we do not intend to get into the business of developing 'our own' codes or of putting our stamp of approval on any of the previously mentioned codes, but instead want to provide a mechanism to help interested researchers obtain their own `stand-alone' and (hopefully) up-to-date versions of these important astrophysical tools. We hope to maintain accurate tracking of which versions of these codes are available at the HEASARC as the codes are modified, debugged, and/or enhanced. We will provide all of the relevant on-line documentation on the codes that we have available from their original authors.

Examples of what is presently available at the HEASARC are: (i) a recent (June 1990) version of the Raymond and Smith (RS) code that was received from John Raymond and installed in February 1991; and (ii) a recent version of the Landini and Monsignori-Fossi (LMF) code that was obtained from Brunella Monsignori-Fossi and installed in July 1991. These are both stand-alone codes written in FORTRAN that, given an input set of plasma parameters and a desired energy range and resolution, calculate the resultant X-ray/EUV spectrum produced by such a plasma.

Such stand-alone plasma emission codes are optimized more for computing and comparing theoretical spectra than for comparison with observed spectra. To accomplish the latter task, most astronomers use customized analysis packages. For example, the widely used X-ray spectral analysis package XSPEC is supported and distributed by the HEASARC. (It can either be run interactively on the HEASARC computer NDADSA, or obtained via tape or electronically from the HEASARC and run on the user's home computer.) As two of its model spectrum options, XSPEC has models "raymond" and "mewe" which are essentially lookup tables constructed from output of the respective codes. The present version of XSPEC installed on NDADSA (Version 7.1) incorporates a somewhat older version of the RS code than that which is available in stand-alone form. The next public release of XSPEC (version 8.0) will include an update of its raymond model based on tables generated from the June 1990 version. This and future releases of XSPEC will have the ability to access both the current and previous versions of the plasma emission models, so as to give the user a better idea of the degree to which the derived plasma parameters depend on the particular version of the plasma code with which his or her data were compared.

Future Work at the HEASARC on Plasma Emission Codes

We are presently planning the following activities in this general area in the near- to long-term:

(i) A comparison will be made of output spectra from different codes (and different versions of the same code) for a variety of astrophysical conditions so as to give some idea as to their inherent accuracy and possible limitations. The results of this comparison will be available in the on-line documentation when they are completed.

(ii) A sample of high-resolution SMM X-Ray Polychromator spectra of solar flares and active regions will be installed for comparison with model spectra generated by the various plasma emission codes.

(iii) Additional thermal (or 'coronal') codes will be installed as needed or requested; for instance, we are hoping to obtain an on-line version of the MG code.

(iv) We will explore whether it is feasible to install more sophisticated plasma emission codes such as time-dependent codes (e.g., Shapiro and Moore 1976; Shull 1982 and/or X-ray photo-ionization codes like HULLAC, Kallman's code, etc.) given our present manpower constraints and their availability for general distribution.

How to obtain copies of the RS and LMF codes from the HEASARC

These codes are of course the end products of an extraordinary (both in quantity and in quality) amount of work by their respective authors. As a public service they have generously made them available for use in the community at large. Scientists who are interested in obtaining their own copies of these codes can either contact their authors directly, or obtain them from the HEASARC.

Comments, questions and requests for the plasma codes at the HEASARC should be addressed to Steve Drake, Code 668, NASA/GSFC, Greenbelt MD 20771 (TEL: 301 286 6962), or by NSI/Science Internet to NDADSA::DRAKE.

References

Gaetz, T.J. and Salpeter, E.E. 1983, Ap. J. Suppl, 52, 155.

Gronenschild, E.H.B.M. and Mewe, R. 1978, Astr. Ap. Supp, 32, 283.

Jordan, C. 1969, MNRAS, 142, 501.
Kato, T. 1976, Ap. J. Supp, 30, 397.
Landini, M. and Monsignori-Fossi, B.C. 1970, Astr. Ap, 6, 468.
------- 1972, Astr. Ap. Supp, 7, 291.
------- 1985, Ap. J, 289, 709.
------- 1990, Astr. Ap. Supp, 82, 229.
Mewe, R. 1972, Solar Phys, 22, 459.
------- 1975, Solar Phys, 44, 383.
Mewe, R. and Gronenschild 1981, Astr. Ap. Supp, 45, 11.
Mewe, R. et al. 1985, Astr. Ap. Supp, 62, 197.
------- 1986, Astr. Ap. Supp, 65, 511.
Raymond, J.C. 1988, in Hot Thin Plasmas in Astrophysics (ed. R. Pallavicini), p. 3.
Raymond, J.C., Cox, D.P., and Smith, B.W. 1976, Ap. J., 204, 290.
Raymond, J.C. and Smith, B.W. 1977, Ap. J. Supp, 35, 419.
Shapiro, P.R. and Moore, R. 1976, Ap. J, 207, 460.
Shull, J.M. 1981, Ap. J. Supp, 46, 27.
------- 1982, Ap. J, 262, 308.
Stern, R. et al. 1978, Ap. J. Supp, 37, 195.
Summers, H.P. 1974, MNRAS, 169, 663.
Tucker, W.H. and Koren, M. 1971, Ap. J, 168, 283.

Proceed to the next article

Return to the previous article

Select another article