1: %
2: % REF: B/
3: % Title: Electron impact excitation of helium-like oxygen up to
4: % n = 4 levels including radiation damping
5: % Authors: Franck Delahaye and Anil K. Pradhan
6: %
7: %
8: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
9: %%% %%%
10: %%% INSTITUTE OF PHYSICS PUBLISHING %%%
11: %%% %%%
12: %%% IOPLPPT.STY LaTeX preprint style file for IOP Journals %%%
13: %%% %%%
14: %%% Copyright 1993 IOP Publishing Ltd %%%
15: %%% %%%
16: %%% Permission is hereby given to use this file for %%%
17: %%% material to be submitted to or published by %%%
18: %%% Institute of Physics Publishing %%%
19: %%% %%%
20: %%% Version 1.1 20 September 1993 %%%
21: %%% %%%
22: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23: %
24: % First we have a character check
25: %
26: % ! exclamation mark " double quote
27: % # hash ` opening quote (grave)
28: % & ampersand ' closing quote (acute)
29: % $ dollar % percent
30: % ( open parenthesis ) close paren.
31: % - hyphen = equals sign
32: % | vertical bar ~ tilde
33: % @ at sign _ underscore
34: % { open curly brace } close curly
35: % [ open square ] close square bracket
36: % + plus sign ; semi-colon
37: % * asterisk : colon
38: % < open angle bracket > close angle
39: % , comma . full stop
40: % ? question mark / forward slash
41: % \ backslash ^ circumflex
42: %
43: % ABCDEFGHIJKLMNOPQRSTUVWXYZ
44: % abcdefghijklmnopqrstuvwxyz
45: % 1234567890
46: %
47: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
48: %
49: \documentstyle[12pt,epsfig]{ioplppt}
50: %\documentstyle[12pt,epsfig]{iopart}
51: %
52: % submitted to Journal of Physics B: Atomic, Molecular and Optical Physics
53: %
54: % Ref: B/
55: %
56: %
57: %
58: \def\las{\mathrel{\hbox{\rlap{\hbox{\lower4pt\hbox{$\sim$}}}\hbox{$<$}}}}
59: \def\gas{\mathrel{\hbox{\rlap{\hbox{\lower4pt\hbox{$\sim$}}}\hbox{$>$}}}}
60: \def\Gamr{\mit\Gamma\sp{\rm r}}
61: \def\Gama{\mit\Gamma\sp{\rm a}}
62: \def\o{\sp{\rm o}}
63: \def\etal{{\it et \thinspace al.}\ }
64: %
65: \begin{document}
66: \jl{2}
67: %
68: \title{Electron Impact Excitation of Helium-like
69: Oxygen up to n=4 levels including radiation damping}
70: %[Electron impact excitation of Fe$^{5+}$]
71:
72: \author{Franck Delahaye and Anil K. Pradhan}
73:
74: \address{Department of Astronomy,
75: Ohio State University \\ Columbus, Ohio,
76: USA, 43210}
77:
78: \begin{abstract}
79:
80: The primary X-ray diagnostic lines in He-like ions
81: are mainly excited by electron impact from the ground
82: level to the $n = 2$ levels, but at high temperatures $n > 2$ levels
83: are also excited. In order to describe the atomic processes more completely
84: collision strengths are computed for O~VII including for the first time
85: all of the following:
86: (i) relativistic fine structure, (ii) levels up to the $n=4$, and (iii)
87: radiation damping of autoionizing resonances.
88: The calculations are carried out using the
89: Breit-Pauli R-matrix (BPRM) method with a 31-level eigenfunction expansion.
90: Resonance structures in collision strengths are delineated in detail up to
91: the $n = 4$ thresholds. For highly charged He-like ions radiation damping
92: of autoionizing resonances is known to be significant.
93: We investigate this effect in detail and find that while resonances
94: are discernibly damped radiatively as the series limit $n \rightarrow \infty$ is
95: approached from below, the overall effect on effective cross sections and
96: rate coefficients is found to be very small. Collision strengths for the
97: principal lines important in X-ray plasma diagnostics, w,x,y and z,
98: corresponding to the 4 transitions to
99: the ground level $1s^2 \ (^1S_0)
100: \longleftarrow 1s2p (^1P^o_1), 1s2p (^3P^o_2), 1s2p (^3P^o_1), 1s2s
101: (^3S_1)$, are explicitly shown. It is found that the effective
102: collision strength of the forbidden z-line is up to a factor of 4 higher
103: at T $< 10^6$ K than previous values. This is
104: likely to be of considerable importance in the diagnostics of
105: photoionized astrophysical
106: plasmas. Significant differences are also found with previous works for
107: several other transitions.
108: This work is carried out as part of the Iron Project-RmaX Network.
109:
110: \end{abstract}
111:
112:
113:
114: \pacno{34.80.Kw}
115: \maketitle
116:
117: \submitted
118:
119: \section{Introduction}
120:
121:
122: Helium-like ions provide the most important X-ray spectral diagnostics
123: in high temperature fusion and astrophysical plasmas.
124: The new generation of X-Ray satellites such as the Chandra X-Ray
125: Observatory and the X-Ray Multi-Mirror Mission-Newton provide
126: high resolution spectra of
127: different types of astronomical objects (e.g. Kaastra \etal 2000,
128: Porquet and Dubau 2000, Porquet \etal 2001). The high sensitivity of these
129: observatories and the high quality of the spectra they produce requires
130: highly accurate atomic data for a precise interpretation.
131: The aim of the Iron Project-R-matrix calculations for X-ray
132: spectroscopy (IP-RmaX) is to calculate extended sets of accurate collision
133: strengths and rate coefficients for all ions of importance in X-Ray
134: diagnostics. Among previous works, the electron impact excitation of
135: Helium like oxygen was
136: previously considered by Pradhan \etal(1981a,b) in the distorted wave
137: and close coupling approximations for transitions up to the $n=2$ levels.
138: Sampson \etal (1983)
139: and Zhang and Sampson (1987) used the Coulomb-Born approximation with
140: exchange, intermediate coupling, and some resonances effects to
141: obtain collision strengths for Helium-like ions, with atomic number
142: Z spanning a large
143: range of values ($4<Z<74$). Kingston and Tayal (1983a,b) calculated the
144: collision strength for two
145: transitions, from the ground state to $2 ^3S_1$ and to $2 ^3P^o_1$, using
146: the close coupling R-matrix (RM) method, and derived the corresponding
147: effective collision
148: strengths. Both, the Pradhan \etal and Kingston and Tayal calculations
149: were in LS coupling. The present work aims at generating a more complete
150: dataset of
151: high reliability for O VII, including all important effects for highly
152: charged ions such as relativistic effects, radiation damping, and
153: resonances in higher complexes up to $n=4$.
154:
155: The method and computations are summerized in section 2.
156: Results for the collision strengths and important issues
157: are discussed in section 3, and the present results for the effective
158: (Maxwellian averaged) collision strengths are
159: compared with previous calculations. The main conclusions are given in
160: section 4, together with an estimate of accuracy of the results.
161:
162: \section{Method and Computations\protect\\}
163:
164: The collisional calculation in the present work has been carried out using
165: the Breit-Pauli R-matrix (BPRM) method as used in the Iron Project (IP) and
166: utilised in a number of previous publications.
167: The aims and methods of the IP are presented in Hummer \etal(1993).
168: We briefly summarise the main features of the method and calculations.
169:
170: In the coupled channel or close coupling (CC) approximation
171: the wave function expansion,
172: $\Psi(E)$, for a total spin and angular symmetry $SL\pi$ or $J\pi$,
173: of the (N+1) electron system
174: is represented in terms of the target ion states as:
175:
176: \begin{equation}
177: \Psi(E) = A \sum_{i} \chi_{i}\theta_{i} + \sum_{j} c_{j} \Phi_{j},
178: \end{equation}
179:
180: \noindent
181: where $\chi_{i}$ is the target ion wave function in a specific state
182: $S_iL_i\pi_i$ or level $J_i\pi_i$, and $\theta_{i}$ is the wave function
183: for
184: the (N+1)$^{th}$ electron in a channel labeled as
185: $S_iL_i(J_i)\pi_i \ k_{i}^{2}\ell_i(SL\pi) \ [J\pi]$; $k_{i}^{2}$ is the
186: incident kinetic energy. In the second sum the $\Phi_j$'s are
187: correlation
188: wave functions of the (N+1) electron system that (a) compensate for the
189: orthogonality conditions between the continuum and the bound orbitals,
190: and (b) represent additional short-range correlations that are often of
191: crucial importance in scattering and radiative CC calculations for each
192: symmetry. The $\Phi_j$'s are also referred to as ``bound channels", as
193: opposed to the continuum or ``free" channels in the first sum over the
194: target states. In the relativistic BPRM calculations the set of
195: ${SL\pi}$
196: are recoupled in an intermediate (pair) coupling scheme
197: to obtain (e + ion) states with total $J\pi$, followed by
198: diagonalisation of the (N+1)-electron Hamiltonian. Details of the
199: diagonalization and the R-matrix method are given in many previous works
200: (e.g. Berrington \etal 1995).
201:
202: The target expansion for the close coupling calculations consists of 31
203: fine-structure levels arising from the 19 LS terms with principal quantum
204: number $n\leq 4$. The target eigenfunctions were developed using the
205: SUPERSTRUCTURE program
206: (Eissner \etal1974) with a version due to Nussbaumer and Storey (1978).
207: The full expansion, together with the scaling
208: factors in the {\it Thomas-Fermi} potential employed in SUPERSTRUCTURE,
209: are given at the end of table 1.
210:
211: In order to estimate the quality of the target wavefunction expansion,
212: we compare the energy levels with those from the {\it National Institue
213: for Standards and Technology} (NIST) in table 1. A better
214: criterion for the accuracy of the wavefunctions is the accuracy
215: of the oscillator strengths for transitions in the target ion. In table 2 we
216: compare the {\it gf}-values with the evaluated compilation from NIST for a
217: number of dipole transitions in O~VII. For the energies the agreement with
218: the NIST values is found to be very good, within 0.05\% for all levels.
219: The oscillator strengths agree well within
220: 10\% (however, for some of the values given by NIST, the estimated accuracy
221: is 30\%). Another accuracy criterion is the level of agreement between the
222: oscillator strengths in the length and the velocity formulations,
223: which we also find to be a few percent for all transitions. The Einstein
224: A-values are also presented in table 2 to enable ready application of
225: the present collisional data in radiative-collisional models for
226: spectral diagnostics (e.g. Porquet and Dubau 2000, Porquet \etal 2001).
227:
228: %-----------------------------------------------------------------
229: \begin{table}
230: \centering
231: \noindent{Table 1: Energy levels compared
232: with observed values (in Rydbergs).}
233: \newline
234: %\label{TabSecInst}
235: \begin{tabular}{llrr}
236: \hline
237: \noalign{\smallskip}
238: Levels & & $E_{calc}$
239: & $E_{obs}$ \\
240: \noalign{\smallskip}
241: \hline
242: \noalign{\smallskip}
243: $1s^2 $ & $^1S{_0}$ & 0.0000 & 0.0000 \\
244: $1s2s$ & $^3S{_1}$ & 41.2438 & 41.2315 \\
245: $1s2p$ & $^3P{_0}^o$ & 41.7933 & 41.7872 \\
246: $1s2p$ & $^3P{_1}^o$ & 41.7942 & 41.7877 \\
247: $1s2p$ & $^3P{_2}^o$ & 41.7997 & 41.7928 \\
248: $1s2s$ & $^1S{_0}$ & 41.8074 & 41.8124 \\
249: $1s2p$ & $^1P{_1}^o$ & 42.2100 & 42.1844 \\
250: $1s3s$ & $^3S{_1}$ & 48.6577 & 48.6509 \\
251: $1s3p$ & $^3P{_0}^o$ & 48.8114 & 48.8044 \\
252: $1s3p$ & $^3P{_1}^o$ & 48.8116 & 48.8044 \\
253: $1s3p$ & $^3P{_2}^o$ & 48.8132 & 48.8044 \\
254: $1s3s$ & $^1S{_0}$ & 48.8217 & 48.8112 \\
255: $1s3d$ & $^3D{_1}$ & 48.8930 & 48.8837 \\
256: $1s3d$ & $^3D{_2}$ & 48.8931 & 48.8842 \\
257: $1s3d$ & $^3D{_3}$ & 48.8935 & 48.8843 \\
258: $1s3d$ & $^1D{_2}$ & 48.8971 & 48.8937 \\
259: $1s3p$ & $^1P{_1}^o$ & 48.9281 & 48.9218 \\
260: $1s4s$ & $^3S{_1}$ & 51.1813 & 51.1798 \\
261: $1s4p$ & $^3P{_0}^o$ & 51.2436 & 51.2360 \\
262: $1s4p$ & $^3P{_1}^o$ & 51.2437 & 51.2360 \\
263: $1s4p$ & $^3P{_2}^o$ & 51.2444 & 51.2360 \\
264: $1s4s$ & $^1S{_0}$ & 51.2475 & 51.2410 \\
265: $1s4d$ & $^3D{_1}$ & 51.2767 & 51.2675 \\
266: $1s4d$ & $^3D{_2}$ & 51.2767 & 51.2600 \\
267: $1s4d$ & $^3D{_3}$ & 51.2769 & 51.2720 \\
268: $1s4f$ & $^3F{_2}^o$ & 51.2786 & 51.2698 \\
269: $1s4f$ & $^3F{_3}^o$ & 51.2785 & 51.2698 \\
270: $1s4f$ & $^3F{_4}^o$ & 51.2787 & 51.2698 \\
271: $1s4d$ & $^1D{_2}$ & 51.2790 & 51.2739 \\
272: $1s4f$ & $^1F{_3}^o$ & 51.2787 & 51.2755 \\
273: $1s4p$ & $^1P{_1}^o$ & 51.2916 & 51.2870 \\
274: \noalign{\smallskip}
275: \hline\\
276: \end{tabular}
277: \\
278: Spectroscopic configurations: $1s^2$, 1s2s, 1s2p
279: 1s3s, 1s3p, 1s3d, 1s4s, 1s4p, 1s4d, 1s4f.\\
280: Correlation configurations: $2s^2$, $2p^2$, 2s2p
281: 2s3s, 2s3p, 2s3d, 2s4s, 2s4p, 2s4d, 2s4f.\\
282: Scalling factors: $\lambda_{1s}$ = 0.9932,
283: $\lambda_{2s}$ = 1.0759, $\lambda_{2p}$ = 0.9217, $\lambda_{3s}$ = 1.0306,\\
284: $\lambda_{3p}$ = 0.9023, $\lambda_{3d}$ = 0.9547, $\lambda_{4s}$ = 1.0182,
285: $\lambda_{4p}$ = 0.9038, $\lambda_{4d}$ = 0.9512,$\lambda_{4f}$ = 1.0200.\\
286: \end{table}
287:
288:
289: %--------------------------------------------------------------------------
290: %--------------------------------------------------------------------------
291:
292: \begin{table}
293: \noindent{Table 2: Comparison of Aij and g*f
294: values compiled by NIST and thoses
295: obtained with SUPERSTRUCTURE (calc).\\ \\}
296: %\label{TabSecInst}
297: \begin{tabular}{ccrrrrrr}
298: \hline
299: \noalign{\smallskip}
300: $Conf_i-Conf_k$ &$Term_i-Term_k$ & $E_i\ (Ryd)$
301: & $E_k\ (Ryd)$ & $A_{ki}^{NIST}\ (s)$ & $A_{ki}^{cal}\ (s)$
302: & $g*f_{NIST}$ & $g*f_{cal}$ \\
303: \noalign{\smallskip}
304: \hline
305: \noalign{\smallskip}
306: $ 1s2-1s2p$& $ ^1S_0$ - $^1P^o_1$& 0.0000 & 42.1844 & 3.309e+12 &3.403e+12 &6
307: .945e-01 &7.142e-01 \\
308: $ 1s2-1s3p$&$ ^1S_0$ - $^1P^o_1 $& 0.0000 & 48.9218 & 9.365e+11 &1.004e+12 &1
309: .462e-01 &1.566e-01 \\
310: $ 1s2s-1s2p$&$ ^3S_1$ - $^3P^o_0$& 41.2315 & 41.7872 & 7.797e+07 &8.058e+07 &3
311: .143e-02 &3.249e-02 \\
312: $ 1s2s-1s2p$&$ ^3S_1$ - $^3P^o_1$& 41.2315 & 41.7877 & 7.820e+07 &8.083e+07 &9
313: .441e-02 &9.757e-02 \\
314: $ 1s2s-1s2p$&$ ^3S_1$ - $^3P^o_2$& 41.2315 & 41.7928 & 8.033e+07 &8.309e+07 &1
315: .587e-01 &1.642e-01 \\
316: $ 1s2p-1s3s$&$ ^3P^{o}_0$ - $ ^3S_1$& 41.7872 & 48.6509 & 2.505e+09 &2.237e+09
317: &1.986e-02 &1.774e-02
318: \\
319: $ 1s2p-1s3d$&$ ^3P^{o}_0$ - $ ^3D_1$& 41.7872 & 48.8837 & 8.982e+10 &8.927e+10
320: &6.662e-01 &6.621e-01
321: \\
322: $ 1s2p-1s3s$&$ ^3P^{o}_1$ - $ ^3S_1$& 41.7877 & 48.6509 & 7.512e+09 &6.739e+09
323: &5.957e-02 &5.344e-02
324: \\
325: $ 1s2p-1s3d$&$ ^3P^{o}_1$ - $ ^3D_1$& 41.7877 & 48.8837 & 6.735e+10 &6.699e+10
326: &4.996e-01 &4.969e-01
327: \\
328: $ 1s2p-1s3d$&$ ^3P^{o}_1$ - $ ^3D_2$& 41.7877 & 48.8842 & 1.213e+11 &1.205e+11
329: &1.499e+00 &1.490e+00
330: \\
331: $ 1s2p-1s3s$&$ ^3P^{o}_2$ - $ ^3S_1$& 41.7928 & 48.6509 & 1.249e+10 &1.131e+10
332: &9.917e-02 &8.983e-02
333: \\
334: $ 1s2p-1s3d$&$ ^3P^{o}_2$ - $ ^3D_1$& 41.7928 & 48.8837 & 4.481e+09 &4.466e+09
335: &3.328e-02 &3.318e-02
336: \\
337: $ 1s2p-1s3d$&$ ^3P^{o}_2$ - $ ^3D_2$& 41.7928 & 48.8842 & 4.033e+10 &4.012e+10
338: &4.992e-01 &4.966e-01
339: \\
340: $ 1s2p-1s3d$&$ ^3P^{o}_2$ - $ ^3D_3$& 41.7928 & 48.8843 & 1.613e+11 &1.608e+11
341: &2.795e+00 &2.787e+00
342: \\
343: $ 1s2s-1s2p$&$ ^1S_0$ - $^1P^o_1$& 41.8124 & 42.1844 & 2.514e+07 &2.509e+07 &6
344: .786e-02 &6.773e-02 \\
345: $ 1s2s-1s3p$&$ ^1S_0$ - $^1P^o_1$& 41.8124 & 48.9218 & 5.055e+10 &5.209e+10 &3
346: .735e-01 &3.849e-01 \\
347: $ 1s2s-1s3p$&$ ^1S_0$ - $^1P^o_1$& 41.8124 & 48.9218 & 5.055e+10 &5.209e+10 &3
348: .735e-01 &3.849e-01 \\
349: $ 1s2p-1s3s$&$ ^1P^{o}_1$ - $ ^1S_0$& 42.1844 & 48.8112 & 2.008e+10 &2.223e+10
350: &5.692e-02 &6.303e-02
351: \\
352: $ 1s2p-1s3d$&$ ^1P^{o}_1$ - $ ^1D_2$& 42.1844 & 48.8937 & 1.523e+11 &1.540e+11
353: &2.106e+00 &2.130e+00
354: \\
355: $ 1s3p-1s3d$&$ ^3P^{o}_0$ - $ ^3D_1$& 48.8044 & 48.8837 & 6.114e+05 &6.200e+05
356: &3.632e-02 &3.684e-02
357: \\
358: $ 1s3p-1s3d$&$ ^3P^{o}_1$ - $ ^3D_1$& 48.8044 & 48.8837 & 4.585e+05 &4.649e+05
359: &2.723e-02 &2.762e-02
360: \\
361: $ 1s3p-1s3d$&$ ^3P^{o}_1$ - $ ^3D_2$& 48.8044 & 48.8842 & 8.426e+05 &8.535e+05
362: &8.236e-02 &8.338e-02
363: \\
364: $ 1s3p-1s3d$&$ ^3P^{o}_2$ - $ ^3D_1$& 48.8044 & 48.8837 & 3.057e+04 &3.099e+04
365: &1.816e-03 &1.841e-03
366: \\
367: $ 1s3p-1s3d$&$ ^3P^{o}_2$ - $ ^3D_2$& 48.8044 & 48.8842 & 2.809e+05 &2.841e+05
368: &2.746e-02 &2.775e-02
369: \\
370: $ 1s3p-1s3d$&$ ^3P^{o}_2$ - $ ^3D_3$& 48.8044 & 48.8843 & 1.127e+06 &1.143e+06
371: &1.539e-01 &1.560e-01
372: \\
373: $ 1s3s-1s3p$&$ ^1S_0$ - $^1P^o_1$& 48.8112 & 48.9218 & 3.864e+06 &3.958e+06 &1
374: .180e-01 &1.210e-01 \\
375: $ 1s3d-1s3p$&$ ^1D_2$ - $^1P^o_1$& 48.8937 & 48.9218 & 7.410e+04 &8.082e+04 &3
376: .508e-02 &3.832e-02 \\
377: \noalign{\smallskip}
378: \hline
379: \end{tabular}
380: \end{table}
381:
382: %-----------------------------------------------------------------
383:
384: The 19 LS terms are recoupled in the relativistic BPRM calculations
385: into the corresponding 31 fine structure
386: levels up to the $n=4$ complex using the routine
387: RECUPD that performs intermediate coupling
388: operations including the one-body Breit-Pauli operators (Hummer et. al.
389: 1993). The reconstructed target eigenfunctions and the
390: resulting target energies reproduce to $10^{-5}$ Ryd the results from
391: SUPERSTRUCTURE, verifying that the algebraic operations have been
392: carried out self-consistently and without loss of accuracy.
393: The collision strengths have been calculated for electron
394: energies $0 \leq E \leq 200$ Ryd. This wide energy range ensures a good
395: coverage of the region where resonances up to the $n=4$ complex are important,
396: as well as the higher energy region where no resonance have been included
397: (all channels are open) but where
398: the background collision strengths still make a significant contribution
399: to the Maxwellian averaged rate coefficient for electron temperatures
400: of interest.
401:
402: The inner region R-matrix basis set included 50 orbitals per angular
403: momentum.
404: Because of the importance of the near threshold resonances in the Maxwellian
405: average rate coefficient, careful attention has been devoted to the
406: resolution and a precise mesh has been chosen. A mesh of $10^{-4}$ Ryd was
407: selected for the region where resonances are important, and a coarser mesh for
408: the region where all channel are open.
409: We included the contribution to the collision strengths from all symmetries
410: with total angular momentum J and both odd and even parities,
411: $J\pi \leq (\frac{35}{2})^{o,e}$. The contribution of higher partial waves
412: was included
413: using the Coulomb-Bethe approximation via the `top-up' facility
414: in the asymptotic region program STGF of the R-matrix package
415: (Burke and Seaton 1986; modified by W. Eissner and G.X. Chen).
416:
417: \section{Results and discussion\protect\\}
418:
419: In figures 1 and 2 we present the collision strengths for transitions from
420: the ground state to levels in the $n=2$ complex, and to levels in the
421: complex $n=3$ respectively. The high resolution of the calculations with
422: a large number of points allow us to resolve clearly all the resonances up to
423: the last threshold in the n=4 complex. We delineate the Rydberg series
424: converging to the different series limits in all three complexes.
425: In both figure 1 and figure 2, the identification of the Rydberg series
426: converging to n=3 and $n=4$ thresholds has been marked.
427: The doubly excited (e + O~VII) $\rightarrow$ O~VI resonance complexes,
428: KMM, KMN etc.. converging
429: towards the different $n=3$ and $n=4$ levels are clearly resolved.
430: We can anticipate from figure 2 that, for some transitions, the low magnitude of
431: the background and the high density of the resonances will make the
432: contribution of these resonances converging to the $n=4$ complex very
433: important and dominant. This is confirmed by our work (discussed
434: later).
435:
436: %----------------------------------------------------------------------------
437: \begin{figure}
438: %\vspace*{-2.0cm}
439: %\centering
440: \psfig{figure=fig1Art.eps,height=17.0cm,width=17.0cm}
441: \vspace*{-1cm}
442: \caption{Collision strengths for transition for the principal lines, z,
443: \ x, y and w (from ground state $1s^2$ $^1S_0$ to $2 ^3S_1,\ 2 ^3P^o_1,
444: \ 2 ^3P^o_2,\ 2 ^1P^o_1$).}
445: \end{figure}
446:
447: \begin{figure}
448: %\vspace*{-2.0cm}
449: %\centering
450: \psfig{figure=fig2Art.eps,height=17.0cm,width=17.0cm}
451: \vspace*{-1cm}
452: \caption{ Collision strength for transition
453: from the ground state to the complex with principal quantum
454: number n=3.
455: $1s^2\ ^1S_0 \rightarrow 3 ^3S_1,\ 3 ^3P^o_1,\ 3 ^3P^o_2,$
456: $\ 3 ^3D_1,\ 3 ^3D_2,\ 3 ^3D_3,\ 3 ^1D_2,\ 3 ^1P^o_1$}
457: \end{figure}
458:
459:
460: %----------------------------------------------------------------------------
461: \subsection{Radiation Damping}
462: It has been previously shown (Presnyakov and Urnov 1979, Pradhan 1981,
463: Pradhan and Seaton 1985), that radiation damping may have significant
464: effect on the resonances in collision strengths for highly charged ions
465: since the radiative decay rates
466: may be large and may compete with autoionization rates, i.e. the effect
467: of dielectronic recombination on electron impact excitation.
468: We studied in detail the radiation damping effect of dielectronic
469: recombination, on resonance structures, collision strengths, and
470: rate coefficients.
471: In figures 3 and 4, we present the Rydberg series converging to the $n=2$
472: and $n=3$ levels coupled to the ground state $1^1S_0$ via strong dipole
473: transitions, $2^1P^o_1$
474: (figure3) and $3^1P^o_1$ (figure4). Since the autoionization rates decrease
475: as $n^{-3}$, and the radiative rate remains constant, radiation damping
476: increases with $n$ and the resonances are wiped out as the series
477: limit is reached. We illustrate the effect for one total (e + ion) symmetry
478: $J\pi = 1^e$. It can be seen how effective the diminishing of resonances
479: is as we approach the threshold. The overall effect of damping, on the averaged
480: collision strength, can be up to a factor two (figure5). However, the region
481: where the effect is important is very small, just below the threshold of
482: convergence. It is called the 'quantum defect region', since we use the Bell
483: and Seaton (1985) (see also Pradhan and Seaton 1985) multi-channel quantum defect theory in this region.
484: In figures 4 and 5 this region corresponds to $\Delta E=0.36$ Ryd
485: ($\nu_{min} = 10$). Overall however we find that
486: for O~VII the effective collision strengths and the rate
487: coefficients are not significantly affected (figure 6 and 7). Indeed, in
488: figure 6, damped and undamped effective collision strength $\Upsilon$
489: curves are indistinguishable (solid line).
490:
491: Figures 4 and 5 show that below threshold the collision strength is constant.
492: This is due to the high value of the effective quantum number reached in
493: our calculations ($\nu \approx 100$), sufficient to illustrate
494: graphically the effect of radiation damping up to the region
495: where the resonances are almost completely damped.
496: In order to resolve all resonances converging to the different thresholds of
497: interest, we used an $\nu$-mesh with about 1000 points for each
498: interval $(\nu\ ,\ \nu +1)$. In the quantum defect region the Coulomb
499: potential dominates the scattering process.
500: In order to directly demonstrate the effect of radiation damping in figures
501: 4 and 5, we
502: show that the collision strengths converge toward the background value
503: calculated neglecting the long-range non-dipole potentials.
504: However, just above the threshold the potential is not only Coulombic
505: but multipole contributions are also important. In order to isolate
506: the effect we switched-off the multipole contributions so
507: that there is continuity across the threshold and only the radiation
508: damping effect is illustrated.
509:
510: %-----------------------------------------------------------------------
511: \begin{figure}
512: %\vspace*{-2.0cm}
513: \centering
514: \psfig{figure=fig3Art.eps,height=17.0cm,width=17.0cm}
515: \vspace*{-1cm}
516: \caption{Partial Collision strength for transition from
517: ground state $1s^2$ $^1S_0$ to $2 ^3S_1,\ 2 ^3P^o_1,\ 2 ^3P^o_2,\ 2 ^1S_0$.
518: dashe line: resonance for $J\pi=1_{even}$ converging to $2 ^1P^o_1$;
519: solid line: resonance for $J\pi=1_even$ converging to $2 ^1P^o_1$ damped
520: by recombination.}
521: \end{figure}
522:
523: \begin{figure}
524: %\vspace*{-2.0cm}
525: \centering
526: \psfig{figure=fig4Art.eps,height=17.0cm,width=17.0cm}
527: \vspace*{-1cm}
528: \caption{Partial Collision strength for transition from
529: ground state $1s^2$$^1S_0$ to $3 ^3S_1,\ 3 ^3P^o_1,\ 3 ^3P^o_2,\ 3 ^1P^o_1$
530: dashe line: resonance for $J\pi=1_{even}$ converging to $3 ^1P^o_1$;
531: solid line: resonance for $J\pi=1_{even}$ converging to $3 ^1P^o_1$ damped
532: by recombination.}
533: \end{figure}
534:
535: \begin{figure}
536: %\vspace*{-2.0cm}
537: \centering
538: \psfig{figure=fig5Art.eps,height=17.0cm,width=17.0cm}
539: \vspace*{-1cm}
540: \caption{Total averaged Collision strength for transition from
541: ground state $1s^2$$^1S_0$ to $2 ^3S_1,\ 2 ^3P^o_1,\ 2 ^3P^o_2,\ 2 ^1S_0$
542: between $2 ^1S_0$ and $2 ^1P^o_1$ thresholds.
543: dashed line: No damping; solid line: Damping effect}
544: \end{figure}
545:
546:
547: %-----------------------------------------------------------------------------
548: \subsection{Effective Collision Strengths}
549:
550: The Maxwellian averaged collision strengths
551:
552: \begin{equation}
553: \Upsilon (T) = \int ^{\infty}_{0}
554: \Omega_{ij}(\epsilon_{j}) e^{-\epsilon_{j}/kT} d(\epsilon_{j}/kT),
555: \end{equation}
556:
557: have been computed for all transitions among levels up to the $n=4$.
558: As has been shown for He-like Fe XXV (Kimura et
559: al 1999, 2000, and Machado-Pelaez et. al. 2001),
560: the resonances arising from the complex $n=N+1$
561: have a strong effect on transitions to the complex $n=N$.
562: Along with the full calculation including all the 31 fine structure states up
563: to $n=4$, we considered a smaller target model, including all 17 levels
564: up to $n=3$ to compare the effect of the resonances.
565: figures 6, 7 and 8 show the effective collision strengths for
566: transitions from the ground state to the $n=2$ levels, as well as
567: transitions among levels of the $n=2$ complex, and transitions from the
568: ground to levels of the $n=3$ complex. We find that for some
569: transitions the
570: contribution from the higher $n=4$ resonances gives rise to a factor of two
571: increase in the
572: effective collision strengths at the temperature of maximum abundance of O~VII
573: ($T \approx 2 \times \ 10^{6}$).
574:
575: %-----------------------------------------------------------------------------
576: \begin{figure}
577: \centering
578: \psfig{figure=fig6Art.eps,height=17.0cm,width=17.0cm}
579: \caption{Effective Collision Strengths for the principal lines (z, x, y,
580: w).
581: solid line: Present work (long dashes on 6b: x and y; solid =(x+y))
582: dot-dashe line: Kingston and Tayal (1983)
583: short-dashe line: Zhang and Sampson (1987)
584: Crosses: Pradhan \etal (1981)
585: Note: The two curves with and without radiation damping effect are
586: indistinguishable (solid line).}
587: \end{figure}
588:
589: %%-----------------------------------------------------------------------------
590: \begin{figure}
591: \centering
592: \psfig{figure=fig7Art.eps,height=17.0cm,width=17.0cm}
593: \caption{Effective Collision Strengths for transitions within the $n=2$
594: complex.
595: solid line: Present work
596: dashed line: Zhang and Sampson (1987)
597: crosses: Pradhan \etal (1981)}
598: \end{figure}
599: %%-----------------------------------------------------------------------------
600: \begin{figure}
601: \centering
602: \psfig{figure=fig8Art.eps,height=17.0cm,width=17.0cm}
603: \caption{Effective Collision Strengths for transitions from $1s^2\ ^1S_0$ to
604: the $n=3$ complex.
605: solid line: Present work up to $n=4$
606: dashed line: Present work up to $n=3$
607: Note: The effect of resonance arising from $n=4$ complex may have a
608: major effect
609: on transition
610: to the $n=3$ complex (a factor 2 for 1 $^1S_0 - 3 ^3D_1$)}
611: \end{figure}
612:
613: %-----------------------------------------------------------------------------
614:
615: We compared the effective collision strengths with previous calculations
616: for the principal lines w, x, y, z, (figure 6a 6b and 6d)corresponding to the
617: 4 transitions to
618: the ground level $1s^2 \ (^1S_0) \longleftarrow 1s2p (^1P^o_1), 1s2p
619: (^3P^o_2), 1s2p (^3P^o_1), 1s2s (^3S_1)$ respectively, that are of primary
620: interest in X-ray spectral diagnostics (e.g. Gabriel and Jordan 1969,
621: Pradhan 1982, Porquet et. al. 2001). Some other transitions are also
622: compared for which the data are available in literature. Illustrative
623: results are
624: presented for the four principal lines (figure 6) as well as four other transitions
625: within the $n=2$ complex (figure 7).
626:
627: Generally the agreement between the different calculations, depending on the
628: transition and temperature, is between 10 - 30\%. However, for some transitions
629: (including the important z-line transition) the differences are larger
630: in some temperature ranges. Basically, theses differences stem from
631: (i) the coupling effects due to the $n=3$ and $n=4$ levels in the close
632: coupling expansion, (ii) relativistic
633: effects included through the Breit-Pauli approximation, (iii)
634: improved delineation of resonances with high-resolution, and (iv)
635: ensuring convergence with complete `top-up' of partial waves. All of
636: these four factors are important in determining the final effective
637: collision strength.
638:
639: In figure 6 we present the transition from the ground level to the
640: different fine structure levels with $n=2$.
641: The z-line (figure 6a) presents the biggest difference with previous
642: calculations,
643: especially at low temperatures. There is a factor 2 difference
644: with Pradhan \etal (1981, crosses), and between 30 \% and a factor of 2 with
645: Kinston
646: and Tayal (1983, dot-dashes) for $T < 2\ 10^{5}$ K, both lying below the
647: present values. Above $10^{5}$ K the difference is within 10\%.
648: The most striking difference however is with the results of
649: Zhang and Sampson (1987, short-dashes) that are much lower and start to
650: converge only at very high temperatures. One might ascribe it to the
651: fact that resonances are not fully considered in their work.
652:
653: The primary cause of differences with previous works is the resolution of
654: complexes of resonances near threshold, {\it in between the $n=2$ levels},
655: where high resolution
656: is crucial and detailed fine structure plays an important role. The
657: $\Omega_{ij}$ at low energies determines the effective collision
658: strength at low temperatures since
659: $\Upsilon _{ij} (T \rightarrow 0) = \Omega _{ij} (E \rightarrow 0) $.
660: The importance of the resonances for this transition was already shown by
661: Pradhan \etal (1981) and Kinston and Tayal (1983a,b) in their LS coupling
662: calculations, both of whom obtain results significantly higher than
663: Zhang and Sampson (1987). The present results demonstrate the need for a
664: relativistic and high-resolution calculation in order to accurately
665: obtain the excitation rates at low temperatures.
666:
667: With the exception of Zhang and Sampson (1987), the previous works
668: did not consider fine structure. Therefore
669: the effective collision strengths for the $1^1S_0 - 2^3P^o_{2,1}$
670: transitions, corresponding to the forbidden and intercombination lines x
671: and y, are shown individually in figure 6b, as well as added together
672: $\Upsilon (x+y)$. The difference with all others
673: is no more than 20\% at all temperatures.
674:
675: For the dipole allowed transition ($1\ ^1S_0\ -\ 2\ ^1P^o$),
676: corresponding to the w-line,
677: the high temperature rates are within 10\%. But still, the low temperature
678: rates are higher by 20\% than those from Pradhan et. al. (1981, crosses).
679:
680: The strong dipole transitions among the excited $n=2$ levels
681: ($2\ ^3S_1\ -\ 2\ ^3P^o$ and $2\ ^1S_0\ -\ 2\ ^1P^o$) (figures 7c and 7d) are
682: very important in spectral diagnostics calculations since they enable
683: the collisional coupling at high electron densities that affects the
684: x, y, and z lines. This also has implications for the competition
685: between the effect of collisional redistribution among these lines, and
686: photoexcitation by background ultraviolet radiation, if present in the
687: X-ray source (e.g. Porquet \etal 2001). The total multiplet collision strength
688: for this transition differs from earlier works by up to 30\%.
689:
690: \section{Conclusion\protect\\}
691: Some of the general conclusions of the paper are as follows.
692:
693: 1. The most complete close coupling calculation using the Breit-Pauli
694: R-Matrix method has been
695: carried out for helium-like oxygen, including resonances up to $n = 4$
696: levels.
697: Detailed studies of radiation damping indicate that it may have a
698: significant effect on the detailed collision strengths in a small energy
699: region below the threshold(s) of convergence, but not on the effective
700: collision strengths.
701: However, radiation damping is important for higher-Z elements
702: since the transition probabilities increase with Z (Pradhan 1983a,b has
703: found the effect on the z-transition to be 9\% in $\Upsilon$ for Fe~XXV
704: at the temperature of maximum abundance of helium-like iron).
705:
706:
707: 2. It is verified that the effects of coupling and resonances from the $n = N+1$
708: complex play an important role
709: in effective collision strengths for the transitions to the complex
710: $n = N$.
711:
712: 3. The new results for the important z-line transition should significantly
713: affect the analysis of
714: O~VII X-ray spectra from photoionized sources (e.g. active galactic
715: nuclei), where O~VII may be abundant at relatively low temperatures.
716: In collisional ionized (coronal) sources the new
717: results may not affect the theoretically computed line intensities
718: significantly at temperatures close to maximum abundance, but should still
719: do so at lower temperatures.
720: It would be preferable to employ the present data in future
721: collisional-radiative and photoionization models.
722:
723: 4. As all relevant atomic effects in electron-ion collisions have been
724: considered, and resonances have been carefully delineated, we should
725: expect the present results to be of definitive accuracy. Nonetheless, we
726: conservatively estimate the precision to be about 10-20\%.
727:
728: 5. All data will be electronically available from the first author from
729: delahaye@astronomy.ohio-state.edu.
730:
731: The authors would like to thank Dr. Werner Eissner for immense help
732: with many aspects of this work. We also thank Guo-Xin Chen for the
733: new and efficient version of STGF including radiation damping
734: used in these calculations.
735: This work was supported partially by the U.S. National Science Foundation
736: and by NASA.
737: The computational work was carried out on the massively
738: parallel Cray T3E and the vector processor
739: Cray T94 at the Ohio Supercomputer Center in Columbus, Ohio.
740:
741: %***
742: %*** E n d o f p a g e 1 o f g a l l e y - m o d e o u t p u t
743: %***
744:
745: %------------------------------------------------------------------------
746:
747: \section*{References}
748:
749: \def\amp{{\it Adv. At. Molec. Phys.}\ }
750: \def\apj{{\it Astrophys. J.}\ }
751: \def\apjs{{\it Astrophys. J. Suppl. Ser.}\ }
752: \def\apjl{{\it Astrophys. J. (Letters)}\ }
753: \def\aj{{\it Astron. J.}\ }
754: \def\aa{{\it Astron. Astrophys.}\ }
755: \def\aas{{\it Astron. Astrophys. Suppl.}\ }
756: \def\aasup{{\it Astron. Astrophys. Suppl.}\ }
757: \def\adndt{{\it At. Data Nucl. Data Tables}\ }
758: \def\cpc{{\it Comput. Phys. Commun.}\ }
759: \def\jqsrt{{\it J. Quant. Spectrosc. Radiat. Transfer}\ }
760: \def\jpb{{\it Journal Of Physics B}\ }
761: \def\pasp{{\it Pub. Astron. Soc. Pacific}\ }
762: \def\mn{{\it Mon. Not. R. astr. Soc.}\ }
763: \def\pra{{\it Physical Review A}\ }
764: \def\prl{{\it Physical Review Letters}\ }
765: \def\zpds{{\it Z. Phys. D Suppl.}\ }
766:
767: \begin{harvard}
768: \item{}Bell R.H., Seaton M.J., 1985, AdSpR 15, 37
769:
770: \item{}Berrington K.A., Burke P.G., Chang J.J., \etal, 1974, Comput.
771: Phys. Commun. 8, 149
772:
773: \item{}Berrington K.A., Burke P.G., Le Dourneuf M., \etal, 1978, Comput.
774: Phys. Commun. 14, 367
775:
776: \item{}Berrington K.A., Eissner, W. and Norrington, P.H. 1995 \cpc 92 290
777:
778: \item{}Burke P.G., Seaton M.J., 1971, Math. Comput. Phys. 10, 1
779:
780: \item{}Burke P.G., Hibbert A., Robb W.D., 1971, J. Phys. B 4, 153
781:
782: \item{}Burke, V. M. and Seaton, M. J. 1986 \jpb 19 L533
783:
784: \item{}Eissner W., Jones M., Nussbaumer H., 1974, Comput. Phys. Commun. 8,270
785:
786: \item{}Kaastra J.S., Mewe, R., Liedahl, D.A., Komosa, S., and Brinkman, A.C.
787: 2000 \aa 354 L83
788:
789: \item{}Kingston A.E., Tayal S.S., 1983, J. Phys. B 16, 3465
790:
791: \item{}Kingston A.E., Tayal S.S., 1983, J. Phys. B 16, L53
792:
793: \item{}Porquet D, Mewe R, Dubau J, Raassen A J J, and Kaastra J S
794: 2001 \aa 376 1113
795:
796: \item{}Porquet D and Dubau J 2000 \aas 143 495
797:
798: \item{}Pradhan A.K., 1981 \prl 47 79
799:
800: \item{}Pradhan A.K., 1982 \apj 263 477
801:
802: \item{}Pradhan A.K., 1983 \pra 28 2113 (a) 2128 (b)
803:
804: \item{}Pradhan A.K., Seaton M.J., 1985, J. Phys. B 18, 1631
805:
806: \item{}Pradhan A.K., Norcross D.W., Hummer D.G., 1981, Ap. J. 246, 1031
807:
808: \item{}Pradhan A.K., Norcross D.W., Hummer D.G., 1981, Phys. Rev. A 23,619
809:
810: \item{}Presnyakov and Urnov A.M., 1979, J. Phys. B 8, 1280
811:
812: \item{}Sampson D.H., Goett S.J., Clark R.E.H., 1983, Atomic Data Nuclear
813: Data Tables 29,467
814:
815: \item{}Zhang H.L., Sampson D.H., 1987, Ap. J. Supp. Ser. 63, 487
816: \end{harvard}
817: %------------------------------------------------------------------------
818:
819: \enddocument
820:
821: \377\377\366\343~\364\353]\377\343]u\363^\272\333\375t\323\336}\347\3159
822:
823:
824: