1: \documentclass[preprint,floatfix,showpacs,preprintnumbers,amsmath,showkeys,amssymb,aps,pra]{revtex4}
2: %\documentclass[twocolumn,floatfix,showpacs,superscriptaddress,preprintnumbers,amsmath,amssymb,aps]{revtex4}
3:
4: \usepackage{graphicx}
5: \usepackage{dcolumn}
6: \usepackage{bm}
7: %\usepackage{times}
8:
9: \begin{document}
10:
11: \title{Nuclear-resonant electron scattering}
12:
13: \author{Adriana P\'alffy}
14: \email{Palffy@mpi-hd.mpg.de}
15:
16: \author{Zolt\'an Harman}
17: \email{Harman@mpi-hd.mpg.de}
18: \affiliation{Max-Planck-Institut f\"ur Kernphysik, Saupfercheckweg 1,
19: 69117 Heidelberg, Germany}
20:
21: \date{\today}
22:
23: \begin{abstract}
24:
25: We investigate nuclear-resonant electron scattering as occurring in the
26: two-step process of nuclear excitation by electron capture (NEEC)
27: followed by internal conversion. The nuclear excitation and decay are
28: treated by a phenomenological collective model in which nuclear states
29: and transition probabilities are described by experimental parameters.
30: We present capture rates and resonant strengths for a number of heavy
31: ion collision systems considering various scenarios for the resonant
32: electron scattering process. The results show that for certain cases
33: resonant electron scattering can have significantly larger resonance
34: strengths than NEEC followed by the radiative decay of the nucleus. We
35: discuss the impact of our findings on the possible experimental
36: observation of NEEC.
37:
38: \end{abstract}
39:
40: \pacs{34.80.Lx, 34.80.Dp, 23.20.Nx, 23.20.-g}
41:
42: \keywords{electron scattering, nuclear excitation, resonant transitions,
43: highly charged ions}
44:
45: \maketitle
46:
47: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
48:
49: \section{Introduction}
50:
51: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
52:
53: Electron scattering on ions and atoms provides a valuable experimental
54: means in both atomic and nuclear physics, and is also of great interest
55: due to the significance of this process in high-temperature plasmas
56: (see, for instance, Ref.~\cite{Williams} and references therein). The
57: elastic scattering of free and quasifree electrons on energetic ions has
58: been studied both theoretically \cite{Bartschat,Bhalla,Johnson,Shingal})
59: and experimentally
60: \cite{Zouros1,Zouros2,Grabbe,Hagmann,Greenwood,Huber_spectrometer}. A
61: number of studies have been carried out on electron impact excitation
62: \cite{Huber,Janzen,Jiang,Suzuki,Wallbank} and resonant electron
63: scattering \cite{Itoh,Badnell,Kollmar}. The latter corresponds to
64: dielectronic capture, i.e., continuum electron capture by the excitation
65: of a bound electron, followed by the Auger decay of the autoionizing
66: state. Since the Auger rates are involved in both steps of this process,
67: the corresponding cross sections are particularly sensitive to the
68: electron-electron interaction. This feature can be used for studies of
69: the relativistic interaction of electrons in the strongest binding
70: nuclear fields available up to now. As an example, the relative
71: contribution of the Breit current-current interaction to the cross
72: section of resonant excitation on hydrogen-like uranium ions was shown to
73: be approximately twice as large as in the case of dielectronic capture
74: followed by radiative de-excitation \cite{Kollmar}.
75:
76: Elastic electron scattering provides an indispensable tool for surveying
77: the electromagnetic structure of ground and excited states of nuclei.
78: Electron scattering as a nuclear probe has the major advantage that the
79: interaction is electromagnetic and hence well known. As a result, for a
80: specific charge distribution, the elastic electron scattering cross
81: section can be calculated by phase-shift analysis techniques
82: \cite{deVries1,deVries2}. As it will be argued in this paper, nuclear-resonant
83: electron scattering in highly charged ions can even
84: provide information about nuclear transitions and excited
85: states via the process of nuclear excitation by electron capture (NEEC).
86:
87: \begin{figure}[Ht]
88: \begin{center}
89: \includegraphics[width=0.45\textwidth]{neec3.eps}
90: \caption{\label{scheme1}(Color online) NEEC recombination mechanism of a
91: continuum electron into the $K$ shell of an initially bare ion, followed
92: by IC of the excited nucleus. The nuclear transition from the ground
93: state $G$ to the first excited state $E$ is pictured schematically on
94: the right-hand side of each panel.}
95: \end{center}
96: \end{figure}
97:
98: In the resonant process of NEEC, the collision of a highly charged ion
99: with a free electron with matching kinetic energy leads to a resonant
100: capture into an atomic orbital with the simultaneous excitation of the
101: nucleus \cite{Palffy1}. This recombination process was first
102: theoretically proposed by the authors of Ref.~\cite{Goldanskii} in the
103: context of laser-produced plasmas, and is the time-reversed process of
104: internal conversion (IC). Although not yet experimentally observed, NEEC
105: has been an interesting subject after experimental observations of
106: atomic physics processes with regard to the structure of the nucleus
107: have been recently reported, such as bound-state internal
108: conversion~\cite{Carreyre} and its time-reversed process of
109: nuclear excitation by electron transition~\cite{Kishimoto}, or the so-called
110: electronic-bridge process, which can be regarded as bound-state internal
111: conversion accompanied by photon emission~\cite{Kekez85,Kalman91,Kalman01}.
112: Se\-ve\-ral theoretical studies have been made
113: concerning NEEC in plasmas \cite{Goldanskii,Harston} or in solid targets
114: \cite{Cue,Kimball1,Kimball2}.
115:
116: As the electron capture in NEEC results in the excitation of the
117: nucleus, $\gamma$ decay of the nucleus or IC are expected in the second
118: step of the process. Several theoretical aspects of NEEC followed by the
119: radiative decay of the excited nucleus have been recently addressed
120: \cite{Palffy1,Palffy2,Palffy3,lifetime_prolongation}, providing
121: theoretical cross sections and discussing the possible experimental
122: observation of NEEC by detecting the photons emitted in the nuclear
123: $\gamma$ decay. In this paper we would like to draw the attention to a
124: two-step process in which NEEC is followed by IC, resulting in
125: nuclear-resonant electron scattering (NRES), as schematically pictured
126: in Figure~\ref{scheme1}.
127:
128: Our motivation in investigating this electron scattering mechanism is
129: twofold. First, nuclear-resonant electron scattering is far more
130: sensitive to the electron-nucleus interaction than NEEC followed by the
131: radiative decay of the nucleus, due to the presence of the IC rate in
132: each of the two steps of the process. This makes NRES a more suitable
133: candidate for exploring the spectral properties and dynamics of heavy
134: nuclei by the use of experimental methods and facilities primarily
135: developed for atomic physics. Especially, NRES may allow the
136: determination of nuclear transition energies and transition
137: probabilities, the study of atomic vacancy effects on nuclear lifetime
138: \cite{lifetime_prolongation} and population mechanisms of excited
139: nuclear levels. A second aspect concerns the experimental observation of
140: NEEC. Theoretical calculations for NEEC followed by IC or radiative
141: decay of the nucleus occurring in scattering measurements are
142: particularly useful in finding candidate isotopes and transitions
143: suitable for experimental observation. For a number of heavy nuclei, the
144: IC rates for low-lying first excited levels are substantially higher than the
145: radiative decay rates, with the immediate consequence that NRES cross
146: sections and resonance strengths are larger than the corresponding
147: values for NEEC followed by the radiative decay of the nucleus.
148: Furthermore, in storage ring experiments aiming at the observation of NEEC by detecting the recombined ions
(as in the case of, e.g., dielectronic recombination
149: experiments~\cite{Bra02,Bra03}) both nuclear decay channels should be
150: taken into account. The interest for electron scattering and
151: recombination experiments at the present and future storage ring
152: facilities of the GSI Darmstadt \cite{GSIdesign_rep,NIMB} makes nuclear-resonant
153: electron scattering in heavy highly charged ions an important
154: issue for the experimental observation of NEEC.
155:
156: In this paper we theoretically investigate resonant scattering of
157: electrons undergoing NEEC followed by IC in two possible
158: cases. In the first considered scenario, NEEC and IC occur in the same
159: atomic orbital, as presented schematically in Figure~\ref{scheme1}. The
160: continuum electron in the initial and final state has then the same
161: kinetic energy in the center-of-mass reference frame. A second
162: possibility considers the case in which the electron is captured into an
163: excited state. NEEC then leads to a doubly-excited intermediate state
164: $d_1$, as depicted in Figure~\ref{scheme2}. This intermediate state can
165: decay via emission of photons from the electron shell or the nucleus, or,
166: alternatively, via IC. Since x-ray emission associated with the
167: electronic de-excitation is faster than nuclear decay of the low-lying
168: excited nuclear states considered here, a second intermediate state
169: $d_2$ in which the electron is in the ground state is reached. This
170: process was denoted as nuclear excitation by electron capture followed
171: by fast x-ray decay (NEECX) in an earlier
172: work~\cite{lifetime_prolongation}, in analogy to the already established
173: notation for the atomic process of resonant transfer and excitation
174: followed by x-ray emission (RTEX). Following NEECX, in the nuclear decay
175: step IC of the excited nucleus occurs if energetically allowed,
176: resulting in a final state characterized by a continuum electron with
177: different kinetic energy than the one in the initial state. We denote
178: the process of NRES with fast x-ray decay of the captured electron by
179: NRESX, in analogy to NEECX and RTEX. This more complicated three-step
180: process is considered because of the advantages for the experiment
181: observation due to the much broader width of the state the electron is
182: captured into, as it will be discussed in detail in Section~\ref{results}.
183:
184: We present total cross sections and resonance strengths for NRES and
185: compare them with the ones presented in Ref.~\cite{Palffy1,Palffy2} for
186: the case of NEEC followed by the $\gamma$ decay of the nucleus. The
187: total cross section derivation and a brief description of the
188: electron-nucleus interaction matrix elements are given in Section II. The
189: electric and magnetic electron-nucleus interactions are considered
190: explicitly and the nucleus is described by the help of a nuclear
191: collective model \cite{Greiner}. The dynamics of the electrons is
192: governed by the Dirac equation as required in the case of high-$Z$
193: elements. Section III presents the numerical results for NRES cross
194: sections and resonance strengths and discusses issues of the possible
195: experimental observation of NEEC. We conclude with a short summary.
196: Atomic units are used throughout this paper unless otherwise specified.
197:
198: \begin{figure}
199: \begin{center}
200: \includegraphics[width=0.45\textwidth]{neec4.eps}
201: \caption{\label{scheme2} (Color online) NEEC recombination mechanism of a continuum
202: electron into the $L$ shell of an initially bare ion, followed by fast
203: x-ray emission from the electronic decay to the $K$-shell ground state.
204: The process is completed by IC of the excited nucleus. See text for
205: further explanations.}
206: \end{center}
207: \end{figure}
208:
209: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
210:
211: \section{\label{theory} Theoretical formalism }
212:
213: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
214:
215: In this section we present the total cross section for the process
216: of NEEC followed by IC of the excited nucleus, derived by means of
217: a perturbative expansion of the transition operator. We consider the
218: nuclear transition from the ground state to the first excited state with
219: the simultaneous capture of a free electron into a bare ion or an ion
220: with a closed-subshell configuration. For the cases in which the
221: electron capture does not occur in the ground state, the subsequent fast
222: electronic x-ray decay is taken into account.
223:
224: \subsection{Total cross section}
225: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
226:
227: The total cross section for NRES can be written with the help of the
228: perturbation expansion of the transition operator, following our
229: formalism presented in Ref.~\cite{Palffy1}. In order to clearly identify
230: the terms contributing to the process under study, in ~\cite{Palffy1} we
231: introduced Feshbach projection operators that separate the Fock space
232: into subspaces corresponding to the possible initial, intermediate and
233: final states. The initial state of the ion-electron system consisting of
234: the nucleus in its ground state, the free electron, and the vacuum state
235: of the electromagnetic field, can be written as a direct product of the
236: state vectors,
237: %%
238: \begin{equation}\label{psi_i}
239: | \Psi_i \rangle = | N_i I_i M_i , \vec{p}_i m_{s_i}^+,0 \rangle \equiv
240: | N_i I_i M_i\rangle \otimes | \vec{p}_i m_{s_i}^+ \rangle \otimes | 0 \rangle \,.
241: \end{equation}
242: %%
243: Here, $\vec{p}_i$ is the asymptotic initial momentum of the electron, $m_{s_i}$
244: its spin projection, and $N$ denotes the nuclear ground state,
245: characterized by the angular momentum $I_i$ and the magnetic quantum
246: number $M_i$.
247:
248: For the two-step process of NEEC followed by IC of the
249: excited nucleus, where the electron capture occurs into the electronic
250: ground state, the intermediate state $|\Psi_d\rangle$ is given by
251: %%
252: \begin{equation}\label{eq:dstate}
253: | \Psi_d \rangle = | N_d I_d M_d, n_d\kappa_d m_d,0 \rangle \equiv
254: | N_d I_d M_d \rangle \otimes | n_d\kappa_d m_d \rangle \otimes | 0 \rangle \,,
255: \end{equation}
256: %%
257: with $n_d$, $\kappa_d$, and $m_d$ being the principal quantum number,
258: Dirac angular momentum, and magnetic quantum number of the bound
259: one-electron state, respectively. The one-electron state is written in the
260: spherical bispinor form
261: %
262: \begin{equation}
263: \langle \vec{r}| n_d\kappa_d m_d \rangle = \psi_{n_d\kappa_dm_d}
264: (\vec{r})=\left(\begin{array}{c} g_{n_d \kappa_d}(r)\Omega_{\kappa_d}^{m_d}
265: (\theta,\varphi)\\ i f_{n_d \kappa_d}(r)\Omega_{-\kappa_d}^{m_d}(\theta,\varphi)\end{array}
266: \right)\ ,\label{bound}
267: \end{equation}
268: %
269: where the $\Omega_{\kappa_d}^{m_d}$ are the spherical spinors
270: \cite{Eichler} and $\theta$ and $\varphi$ are the polar and azimuthal
271: angles associated with the vector $\vec{r}$, respectively. The excited
272: nuclear state is denoted by $|N_d I_d M_d\rangle$.
273:
274: The final state for the two-step resonant
275: electron scattering on nuclei is then characterized by the nucleus in
276: its ground state and the electron in the continuum,
277: %%
278: \begin{equation}
279: | \Psi_f \rangle = | N_f I_f M_f, \vec{p}_f m_{s_f}^-,0 \rangle \equiv
280: | N_f I_f M_f \rangle \otimes | \vec{p}_f m_{s_f}^- \rangle \otimes | 0 \rangle \, .
281: \end{equation}
282:
283: We denote the energy eigenvalues of the states introduced above as $E_i$, $E_d$ and
284: $E_f$, respectively. Furthermore, the initial and final state continuum electronic wave
285: functions are given in the coordinate space representation as
286: the partial wave expansion~\cite{Eichler}
287: %
288: \begin{equation}\label{partialwave}
289: |\vec{p} m_s^\pm \rangle=\sum_{\kappa m m_l}i^l e^{\pm i\Delta_{\kappa}}
290: Y_{l m_l}^*(\Omega_p)
291: C\left(l\ \frac{1}{2}\ j;m_l\ m_s \ m\right)| E_c \kappa m\rangle\ ,
292: \end{equation}
293: %
294: where $E_c$ is the energy of the continuum electron measured
295: from the ionization threshold, $E_c=\sqrt{p^2c^2+c^4}-c^2$. The
296: orbital angular momentum of the partial wave is denoted by $l$ and the
297: corresponding magnetic quantum number by $m_l$. The $+(-)$ sign of the
298: partial wave phases $\Delta_{\kappa}$ corresponds to the initial (final)
299: free electron and the phases are chosen so that the continuum wave
300: function fulfills the boundary conditions of an incoming (outgoing)
301: plane wave and an outgoing (incoming) spherical wave. The total angular
302: momentum quantum number of the partial wave is $j=|\kappa|-\frac{1}{2}$
303: with its projection $m=m_l+m_s$ and the symbol $C\left(j_1,j_2,j_3;m_1,m_2,m_3\right)$
304: stands for the Clebsch-Gordan coefficient. The partial wave functions are
305: represented as
306: \begin{equation}
307: \langle \vec{r}| E_c \kappa m \rangle =
308: \psi_{E_c \kappa m}(\vec{r})=
309: \left(\begin{array}{c} g_{E_c \kappa}(r)\Omega_{\kappa}^{m}(\theta,\varphi)\\
310: i f_{E_c \kappa}(r)\Omega_{-\kappa}^{m}(\theta,\varphi)\end{array}
311: \right)\ .\label{cont}
312: \end{equation}
313:
314: Following the formalism presented in our previous work~\cite{Palffy1},
315: the total cross section as a function of the total initial energy
316: $E$ for the two-step process $i\to d\to f$ is given by
317: %%
318: \begin{equation}
319: \sigma_{i \to d \to f}(E) = \frac{2\pi^2}{p^2}
320: \frac{A_{\rm IC}^{d \to f} Y_n^{i \to d}}{\Gamma_d} L_d(E-E_d) \, ,
321: \label{tcs1}
322: \end{equation}
323: %%
324: where $A_{\rm IC}^{d \to f}$ is the IC decay rate of the nuclear excited
325: state and $Y_n^{i \to d}$ the NEEC rate. In the denominator, $\Gamma_d$
326: denotes the total natural line width of the nuclear excited state, given
327: by the sum of the partial IC and $\gamma$ decay widths,
328: $\Gamma_d=\Gamma_{\rm IC}+ \Gamma_{\gamma}$. The continuum electron
329: energy dependence is given by the well-known Lorentz line profile
330: function
331: %%
332: \begin{equation}
333: L_d(E_c - E_{\rm exc} - \varepsilon_{n_d \kappa_d}) = L_d(E-E_d) =
334: \frac{\Gamma_d / 2\pi}{(E-E_d)^2 + \frac{1}{4} \Gamma_d^2}
335: \end{equation}
336: %%
337: %centered on the resonance energy $E_d$ and
338: with the width $\Gamma_d$ given by the natural width of the excited
339: nuclear state. Here we introduced the notation $E_{\rm exc}$ for the
340: nuclear excitation energy and $\varepsilon_{n_d \kappa_d}$ for the
341: energy of the bound intermediate electronic state. The cross section
342: formula (\ref{tcs1}) is valid in the resonant case, i.e. for continuum
343: electron energies $E_c$ approximately fulfilling the resonance condition
344: $E_c = E_{\rm exc}+\varepsilon_{n_d \kappa_d}$.
345:
346: Since NEEC is
347: the time-reversed process of IC, and we consider cases involving
348: transitions between two nuclear levels only, the rates of the two processes
349: occurring between the states $d$ and $f$ and $i$ and $d$ can be related
350: by the principle of detailed balance,
351: %%
352: \begin{equation}
353: A^{d \to f}_{\rm IC}=\frac{2(2I_i+1)}{(2I_d+1)(2j_d+1)}Y_n^{i\to d} \, ,
354: \end{equation}
355: %%
356: where $j_d$ denotes the total angular momentum of the bound electron.
357:
358: For the resonant three-step process depicted in Figure~\ref{scheme2}, in
359: which NEEC occurs into an excited electronic state with subsequent fast
360: x-ray emission, the first intermediate state given in
361: Eq.~(\ref{eq:dstate}) becomes
362: %%
363: \begin{eqnarray}
364: | \Psi_{d_1} \rangle &=& | N_d I_d M_d, n_d^*\kappa^*_d m^*_d ,0 \rangle\\
365: &\equiv& | N_d I_d M_d \rangle \otimes |n_d^*\kappa^*_d m^*_d\rangle \otimes
366: |0 \rangle \,, \nonumber
367: \end{eqnarray}
368: %%
369: where $n_d^*$, $\kappa_d^*$ and $m^*_d$ are the quantum numbers of the
370: excited electronic state. The second intermediate state following the
371: fast x-ray electronic decay is characterized by the electron in the
372: ground state, the excited nucleus and a photon with wavenumber $\vec{k}$
373: and polarization $\nu=1,2$,
374: %%
375: \begin{eqnarray}
376: | \Psi_{d_2} \rangle &=& | N_d I_d M_d, n_d\kappa_d m_d ,\vec{k} \nu \rangle\\
377: &\equiv&
378: | N_d I_d M_d \rangle \otimes |n_d\kappa_d m_d\rangle \otimes |\vec{k} \nu \rangle \,.
379: \nonumber
380: \end{eqnarray}
381: %%
382: The photon is emitted in the x-ray decay of the electron from
383: the excited state $| n_d^*\kappa^*_d m^*_d \rangle$ to the ground state
384: $|n_d\kappa_d m_d \rangle$. The projection operator formalism presented
385: in Ref.~\cite{Palffy1} is extended to account for the emission of such a photon.
386: The corresponding total cross section for the three-step process
387: $i \to d_1 \to d_2 \to f$ can be written as
388: %%
389: \begin{equation}
390: \sigma_{i \to d_1 \to d_2 \to f}(E) =\frac{2\pi^2}{p^2}\frac{A_{\rm IC}^{d_2\to
391: f}}{\Gamma_{d_2}}\frac{A_{\rm x-ray}^{d_1\to
392: d_2}}{\Gamma_{d_1}} Y^{i\to d_1}_{n}L_{d_1}(E-E_{d_1})\, ,
393: \label{tcs2}
394: \end{equation}
395: %%
396: where $A_{\rm x-ray}$ is the electronic radiative decay rate and
397: $\Gamma_{d_1}$ and $\Gamma_{d_2}$ are the widths of the two intermediate
398: states. The width $\Gamma_{d_1}$ of the doubly-excited state $d_1$ is
399: given as the sum of the nuclear and electronic widths,
400: $\Gamma_{d_1}=\Gamma_{\gamma}+\Gamma_{\rm IC}+ \Gamma_{\rm x-ray}$, and
401: can be approximated as $\Gamma_{d_1}\simeq\Gamma_{\rm x-ray}$ due to
402: the difference of magnitude of the electronic and nuclear widths. The
403: natural width of the nuclear excited state determines the width of the second
404: intermediate state $\Gamma_{d_2}$. The Lorentz profile is characterized
405: in this case by the width of the first intermediate state,
406: $\Gamma_{d_1}\simeq\Gamma_{\rm x-ray}$. Because of the large lifetime of
407: the nuclear excited state, in Eq.~(\ref{tcs2}) an additional second
408: term standing for the process in which the nuclear decay occurs prior to
409: the electronic decay can be neglected.
410:
411: The integration of the cross section over the continuum electron energy
412: gives the resonance strength $S$ for a given capture process. In the
413: case of the two-step process $i\to d\to f$ described by the total cross
414: section in Eq.~(\ref{tcs1}), the continuum electron momentum $p$ and
415: thus the NEEC rate $Y_n^{i \to d}$ are practically constant in the
416: energy interval defined by the very narrow nuclear width. Since the
417: Lorentz function is normalized to unity, the resonance strength can be
418: written as
419: %
420: \begin{equation}
421: S = \frac{2\pi^2}{p^2}
422: \frac{A_{\rm IC}^{d \to f} Y_n^{i \to d}}{\Gamma_d} \, .
423: \label{rs1}
424: \end{equation}
425: %
426: For the more complicated three-step process involving NEEC into an
427: excited electronic state followed by x-ray emission and IC of the
428: captured electron, the resonance strength is obtained by integrating the
429: total cross section given in Eq.~(\ref{tcs2}), and has the expression
430: %
431: \begin{equation}
432: S = \frac{2\pi^2}{p^2}\frac{A_{\rm IC}^{d_2\to
433: f}}{\Gamma_{d_2}}\frac{A_{\rm x-ray}^{d_1\to
434: d_2}}{\Gamma_{d_1}} Y^{i\to d_1}_{n}
435: \, .
436: \label{rs2}
437: \end{equation}
438: %
439: Similarly, since the transition width $\Gamma_{d_1}$ is still much
440: smaller than the continuum electron energy, we have assumed here
441: resonance values for the momentum $p$ and the NEEC rate $Y_n^{i \to d}$.
442:
443: \subsection{The electron-nucleus interaction}
444: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
445:
446: The NEEC rates in Eqs.~(\ref{tcs1}) and (\ref{tcs2}) are proportional to
447: the squared matrix elements of the electric and magnetic
448: electron-nucleus interactions and have the expression \cite{Palffy1}
449: %%
450: \begin{eqnarray}\label{Yrate}
451: &&Y_n^{i \to d}=\frac{2\pi}{2(2I_i+1)}\sum_{{M_{i}} m_{s_i}}\sum_{M_{d} m_d}
452: \int d\Omega_p \rho_i\\
453: &&
454: \times |\langle N_d I_d M_d,n_d\kappa_d m_d,0|H_{en}+H_{magn}|N_i I_i M_i,\vec{p}_im_{s_i}^+,0\rangle|^2 \, . \nonumber
455: \end{eqnarray}
456: %%
457: Here, the integral is performed over the incoming electron direction
458: $\Omega_p$ and $\rho_i$ denotes the density of the electronic continuum states
459: ($\rho_i=1$ in the units applied here). The electron-nucleus interaction Hamiltonians
460: $H_{en}$ and $H_{magn}$ describe the electric and magnetic transitions of the
461: nucleus, respectively. We adopt the Coulomb gauge for the
462: electron-nucleus interaction $H_{en}$, since it allows the separation of
463: the dominant Coulomb attraction between the electronic and nuclear
464: degrees of freedom,
465: %%
466: \begin{equation}\label{eq:coulomb}
467: H_{en}=\int d^3r_n\frac{\rho_n(\vec{r}_n)}{|\vec{r}_e-\vec{r}_n|}\, .
468: \end{equation}
469: %%
470: Here, $\rho_n(\vec{r}_n)$ is the nuclear charge density and the
471: integration is performed over the whole nuclear volume. The magnetic
472: interaction Hamiltonian accounts for the recombination of the free
473: electron by exchanging a virtual transverse photon. In the limit of
474: long exchange photon wavelength, the magnetic interaction Hamiltonian is
475: approximated by
476: %%
477: \begin{equation}
478: H_{magn} = - \frac{1}{c} \vec{\alpha} \int d^3r_n
479: \frac{\vec{j}_n(\vec{r}_n)}{|\vec{r}-\vec{r}_n|}\, ,
480: \label{hmagn}
481: \end{equation}
482: %%
483: where $\vec{j}_n(\vec{r}_n)$ is the nuclear current vector and $\vec{\alpha}$ is the vector
484: of Dirac matrices.
485:
486: For describing the nucleus we use a collective model \cite{Greiner} in
487: which the excitations of the nucleus are assumed to be vibrations or rotations of the
488: nuclear surface. The expressions of the nuclear charge density and
489: current in the two interaction Hamiltonians can then be written in terms
490: of nuclear collective coordinates by means of a nuclear surface
491: parameterization. Since the details of the calculation of the
492: interaction Hamiltonian matrix elements are given elsewhere
493: \cite{Palffy1,Palffy2}, here we only present their final expressions.
494: The NEEC transition probability per unit time is given by
495: %%
496: \begin{eqnarray}
497: \label{aprate}
498: Y^{(e)}_n&=&\frac{4\pi^2\rho_i}{(2L+1)^2} B (EL,I_i\!\to \!I_d)(2j_d+1)\\
499: &\times&
500: \sum_\kappa |R^{(e)}_{L,\kappa_d,\kappa}|^2\
501: %C\left(j_d\ L\ j;\frac{1}{2}\ 0\ \frac{1}{2}\right)^2\, , \nonumber
502: (2j+1) \left(\begin{array}{ccc} j_d&j&L\\\frac{1}{2}&-\frac{1}{2}&0\end{array}\right)^2
503: \, , \nonumber
504: \end{eqnarray}
505: %%
506: for electric transitions of multipolarity $L$. The quantity
507: %%
508: \begin{equation}
509: B(\lambda L,I_i\!\to \! I_d)=
510: \frac{1}{2I_i+1}|\langle N^* I_d\|\mathcal{M}_L\|NI_i\rangle |^2
511: \end{equation}
512: %%
513: represents the reduced nuclear transition probability, where
514: $\lambda$ stands for electric ($E$) or magnetic ($M$) and $\mathcal{M}$
515: is the corresponding multipole moment operator. We have denoted by
516: $R^{(e)}_{L,\kappa_d,\kappa}$ the electronic matrix element
517: %%
518: \begin{equation}
519: R^{(e)}_{L,\kappa_d,\kappa}=\int_0^\infty drr^{-L+1}\Big(f_{n_d\kappa_d}(r)
520: f_{E_c\kappa}(r)+g_{n_d\kappa_d}(r)g_{E_c\kappa}(r)\Big)\ ,
521: \label{radial1}
522: \end{equation}
523: %%
524: where $g_{E_c\kappa}(r)$ and $f_{E_c\kappa}(r)$ are the large and small radial components of the
525: relativistic continuum electron wave function in Eq.~(\ref{cont}),
526: respectively, and $g_{n_d\kappa_d}(r)$ and $f_{n_d\kappa_d}(r)$ are the corresponding components
527: of the bound Dirac wave functions as in Eq. (\ref{bound}). The last factor
528: in the summand of Eq. (\ref{aprate}) is a 3-$j$ symbol.
529: For magnetic
530: transitions of multipolarity $L$, the NEEC rate has the expression
531: %%
532: \begin{eqnarray}
533: {Y_n^{(m)}}&=&\frac{4\pi^2\rho_i}{L^2(2L+1)^2}B(ML,I_i\!\to \! I_d)(2j_d+1)
534: \nonumber \\
535: &\times&
536: \sum_{\kappa}
537: \left|R^{(m)}_{L,\kappa_d,\kappa}\right|^2
538: (2j+1)(\kappa_d+\kappa)
539: \left(\begin{array}{ccc} j_d&j&L\\\frac{1}{2}&-\frac{1}{2}&0\end{array}\right)^2 ,
540: \end{eqnarray}
541: %%
542: where we introduced the following notation for the radial integral:
543: %%
544: \begin{equation}
545: R^{(m)}_{L,\kappa_d,\kappa}=\int_0^{\infty}dr r^{-L+1}\bigg
546: (g_{n_d\kappa_d}(r)f_{E_c\kappa}(r)+\
547: f_{n_d\kappa_d}(r)g_{E_c\kappa}(r)\bigg) \,.
548: \label{radial2}
549: \end{equation}
550: %%
551: The radial expressions $R^{(e)}_{L,\kappa_d,\kappa}$ and
552: $R^{(m)}_{L,\kappa_d,\kappa}$ are integrated numerically.
553:
554:
555: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
556:
557: \section{\label{results} Numerical results}
558:
559: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
560:
561: We calculate NRES total cross sections and resonance strengths for a
562: number of systems involving highly charged ions of stable or long-lived
563: ground state isotopes. In the first place we investigate the two-step
564: process of NRES in which electron capture occurs into the electronic
565: ground state and consequently the NEEC and IC bound electronic states
566: coincide, as depicted in Figure~\ref{scheme1}. NEEC occurring in bare
567: ions is considered for $E2$ transitions from the $0^+$ ground states to
568: the first $2^+$ excited nuclear states of $^{154}_{64}\mathrm{Gd}$,
569: $^{164}_{66}\mathrm{Dy}$, $^{170}_{68}\mathrm{Er}$,
570: $^{174}_{70}\mathrm{Yb}$, $^{178}_{72}\mathrm{Hf}$ and
571: $^{180}_{74}\mathrm{W}$. The energies of the excited nuclear levels
572: $E_{\rm exc}$ as well as the reduced transition probabilities $B(E2)$,
573: needed for the calculation of the natural width of the nuclear excited
574: state and the NEEC rate, are taken from
575: Ref.~\cite{Raman}. The $M1$ transitions between the ground states and the
576: first excited states of $^{155}_{64}\mathrm{Gd}$,
577: $^{165}_{67}\mathrm{Ho}$, $^{173}_{70}\mathrm{Yb}$,
578: $^{175}_{71}\mathrm{Lu}$, $^{179}_{72}\mathrm{Hf}$,
579: $^{185}_{75}\mathrm{Re}$ and $^{187}_{75}\mathrm{Re}$ are also
580: considered. For these cases, the dominant $M1$ multipolarity is
581: accompanied by a weaker $E2$ component. For the NEEC and IC rates, we
582: take into account the multipole mixing by considering both Hamiltonians
583: $H_{en}$ corresponding to the $E2$ transition and $H_{magn}$
584: corresponding to the $M1$ transition in Eq.~(\ref{Yrate}). Due to the
585: specific parity of the electronic wavefunction components, the mixed
586: terms in Eq.~(\ref{Yrate}) vanish and the NEEC rate can be written as a
587: sum of the partial NEEC rates for the separate $M1$ and $E2$
588: multipolarities. A similar result is obtained for the $\gamma$ decay
589: rate of a mixed multipole transition. The nuclear data for the magnetic
590: transitions were taken from
591: \cite{gd155,ho165,yb173,lu175,hf179,re185,re187}.
592:
593: In the following we compare NRES, i.e., NEEC followed by IC, with the
594: case when the excited nuclear state occurring in NEEC decays
595: radiatively. Resonance strengths for both scenarios involving electronic
596: capture into the $K$ shell of a bare ion are compared in Table~\ref{K}.
597: The resonance strengths are indexed according to the nuclear decay
598: channel as $S_{\rm IC}$ and $S_{\gamma}$. In the case of NEEC followed
599: by the $\gamma$ decay of the nucleus, the resonance strength is given by
600: \cite{Palffy1}
601: %
602: \begin{equation}
603: S_{\gamma} = \frac{2\pi^2}{p^2}
604: \frac{A_{\gamma}^{d \to f} Y_n^{i \to d}}{\Gamma_d} \, ,
605: \end{equation}
606: %
607: where $A_{\gamma}^{d \to f}$ is the nuclear $\gamma$ decay rate, related
608: to the reduced transition probability $B$ and nuclear excitation energy
609: $E_{\rm exc}$ by \cite{Ring}
610: %
611: \begin{equation}
612: A_{\gamma}^{d \to f}=\frac{8\pi(L+1)}{L[(2L+1)!!]^2}\left(E_{\rm exc}\right)^{2L+1}B(\lambda L,I_e\!\to \!I_g)\, .
613: \end{equation}
614: %
615:
616: For the calculation of the NEEC and IC rates, the numerical evaluation
617: of the radial integrals $R_{L,\kappa_d,\kappa}$ [see
618: Eqs.~(\ref{radial1}) and (\ref{radial2})] is needed. We consider
619: Coulomb-Dirac wave functions for the continuum electron and wave
620: functions calculated with the GRASP92 package~\cite{Par96} assuming the
621: potential of a homogeneously charged nucleus for the bound electron.
622: The value of $R_{L,\kappa_d,\kappa}$ is not affected by finite nuclear
623: size effects on the accuracy level of our calculations. Nevertheless,
624: the finite size of the nucleus has a sensitive effect on the energy
625: levels of the bound electron. The energy of the bound electronic state
626: is calculated with GRASP92 and includes one-loop one-electron quantum
627: electrodynamic (QED) terms, and in the case of many-electron bound states
628: approximate QED screening corrections.
629:
630: \begin{table*}[htb]
631: \caption{\label{K} Resonance strength comparison between NRES ($S_{\rm
632: IC}$) and NEEC followed by $\gamma$ decay ($S_{\gamma}$) for various
633: heavy ion collision systems involving capture of the free electron in
634: the $1s_{1/2}$ orbital of bare ions. The nuclear excitation energy
635: $E_{\rm exc}$, the continuum electron energy at resonance in the
636: center-of-mass frame $E_c$, and the multipolarity of the transition $L$
637: are given in the second, third and fourth column, respectively. }
638: \begin{ruledtabular}
639: \begin{tabular}{lrrcrr}
640: Isotope & $E_{\rm exc}$(keV) & $E_{c}$(keV) & $L$ & $S_{\rm IC}$ (b eV)& $S_{\gamma}$ (b eV)\\
641:
642: \hline
643: $^{154}_{64}\mathrm{Gd}$ & 123.071 &64.005 &$E2$ & 1.21$\times 10^{-2}$ & 2.87$\times 10^{-2}$\\
644: $^{164}_{66}\mathrm{Dy}$ & 73.392 &10.318 &$E2$ & 4.93$\times 10^{-2}$ & 3.86$\times 10^{-2}$\\
645:
646: $^{170}_{68}\mathrm{Er}$ & 78.591 &11.350 &$E2$ & 4.90$\times 10^{-2}$ & 4.69$\times 10^{-2}$\\
647:
648: $^{174}_{70}\mathrm{Yb}$ & 76.471 &4.897 &$E2$ & 3.39$\times 10^{-3}$ & 3.61$\times 10^{-3}$\\
649: $^{178}_{72}\mathrm{Hf}$ & 93.180 &17.103 &$E2$ & 3.11$\times 10^{-2}$ & 4.64$\times 10^{-2}$\\
650:
651: $^{180}_{74}\mathrm{W}$ & 103.557 &22.776 &$E2$ & 2.30$\times 10^{-2}$ & 4.41$\times 10^{-2}$\\
652:
653: \hline
654: $^{155}_{64}\mathrm{Gd}$ & 60.008 &0.942 &$M1+E2$ & 8.48 & 2.19 \\
655: $^{165}_{67}\mathrm{Ho}$ &94.700 & 29.563 &$M1+E2$ & 1.19 & 0.88\\
656:
657: $^{173}_{70}\mathrm{Yb}$ & 78.647 & 7.073 &$M1+E2$ & 3.85 & 1.31\\
658: $^{175}_{71}\mathrm{Lu}$ & 113.804 &40.002 &$M1+E2$ & 0.153 & 0.151\\
659: $^{179}_{72}\mathrm{Hf}$ & 122.7909 &46.714 &$M1+E2$ & 0.327 & 0.348\\
660: $^{185}_{75}\mathrm{Re}$ & 125.358 & 42.198 &$M1+E2$ & 1.74 & 1.47\\
661: $^{187}_{75}\mathrm{Re}$ & 134.243 &51.083 &$M1+E2$ & 1.15 & 1.18\\
662: \end{tabular}
663:
664: \end{ruledtabular}
665: \end{table*}
666: %
667:
668: The comparison in Table~\ref{K} shows that the resonance strengths for
669: NRES and for NEEC followed by $\gamma$ emission are typically on the
670: same order of magnitude. For $E2$ transitions, the $\gamma$ decay of the
671: excited nuclear state tends to dominate over the IC decay, so that
672: $S_{\rm IC} <S_{\gamma}$. For magnetic dipole transitions, typically the
673: NRES resonance strength values are larger than the ones for NEEC
674: followed by $\gamma$ emission, culminating with the case of
675: $^{155}_{64}\mathrm{Gd}$, for which $S_{\rm IC}$=8.48~b~eV and
676: $S_{\gamma}$=2.19~b~eV. The difference between $S_{\rm IC}$ and
677: $S_{\gamma}$ is given by the decay channel only, namely, by the decay
678: rates of the nuclear excited state $A_{\rm IC}$ and $A_{\gamma}$,
679: respectively. The ratio $\alpha= A_{\rm IC}/A_{\gamma}$ denotes the IC
680: coefficient, whose values are calculated for a particular bound shell or
681: orbital. The behaviour of $\alpha$ with respect to the capture orbital
682: depends on the multipolarity of the nuclear transition. While for $E2$
683: transitions $p$ orbitals have larger $\alpha$ values, for $M1$
684: transitions the $s$ orbitals have a stronger contribution to the total
685: IC coefficient. It is therefore not surprising that in Table~\ref{K},
686: with resonance strengths considering NEEC into the $1s_{1/2}$ orbital of
687: bare ions, for $M1$ transitions the NRES resonance strengths are typically
688: larger than the ones for NEEC followed by $\gamma$ decay, $S_{\rm
689: IC}\gtrsim S_{\gamma}$. In addition, the NRES resonance strengths for $M1$ transitions are
690: substantially larger than the ones for $E2$ transitions, also due to the broader
691: natural line widths of the former.
692:
693:
694:
695: For heavier even-even nuclei such as the actinides
696: $^{232}_{90}\mathrm{Th}$, $^{236}_{92}\mathrm{U}$,
697: $^{238}_{92}\mathrm{U}$, and $^{248}_{96}\mathrm{Cm}$, the capture into
698: the $K$ shell is not possible since the binding energy of the $1s_{1/2}$
699: electron is larger than the nuclear excitation energy. These nuclei
700: present first-excited $2^+$ states lying at about 40~keV above the $0^+$
701: ground state. Due to their low transition energies and large
702: corresponding IC coefficients, the actinide nuclei are prospective
703: candidates for NRES with recombination into the $L$ shell. In Table~\ref{U_LL}, we consider NEEC with capture
704: into the $2s_{1/2}$, $2p_{1/2}$ and $2p_{3/2}$ orbitals of the ground
705: state electronic configuration of the He-like ($1s_{1/2}^2$), Be-like
706: ($1s_{1/2}^22s^2_{1/2}$) and C-like ($1s_{1/2}^22s_{1/2}^22p_{1/2}^2$) ions,
707: respectively. NRES resonance strengths are compared to the ones of NEEC
708: followed by $\gamma$ decay of the nucleus. Unlike the cases of NEEC
709: occurring in bare ions, in this case the width of the nuclear excited
710: state may also contain terms corresponding to the IC decay of the bound
711: electrons in the initial electronic configuration. The presence of the
712: $K$-shell electrons does not play any role in the nuclear decay, since
713: the low energy of the nuclear transition does not allow their IC. For
714: Be-like and C-like ions, however, the IC decay rates of the $2s_{1/2}$
715: and $2p_{1/2}$ orbital electrons contribute to the total width of the
716: nuclear excited state. For the calculation of the
717: radial wave functions for the continuum electron, we assume a total
718: screening of the nuclear charge, i.e., we use Coulomb-Dirac functions
719: with an effective nuclear charge $Z_{\rm eff} = Z-N$, where $N$ stands
720: for the number of bound electrons. For the bound electron wave
721: functions, the electron-electron interaction is accounted for in the
722: Dirac-Fock approximation.
723:
724: %
725: \begin{table*}[htb]
726: \caption{\label{U_LL} Resonance strength comparison between NRES
727: ($S_{\rm IC}$) and NEEC followed by $\gamma$ decay ($S_{\gamma}$) for
728: several heavy ion collision systems involving capture of the free
729: electron in the $2s_{1/2}$ orbital of He-like ions, the $2p_{1/2}$
730: orbital of Be-like ions and the $2p_{3/2}$ orbital of C-like ions. The
731: nuclear excitation energy $E_{\rm exc}$, the continuum electron energy
732: at resonance in the center-of-mass frame $E_c$, and the capture orbital
733: $nl_j$ are given in the second, third and fourth column, respectively.}
734: \begin{ruledtabular}
735: \begin{tabular}{lrrcrr}
736: Isotope & $E_{\rm exc}$(keV) & $E_{c}$(keV) & $nl_j$ & $S_{\rm IC}$ (b eV)& $S_{\gamma}$ (b eV)\\
737: \hline
738: & &18.244 & $2s_{1/2}$ & 0.011 &$5.44\times 10^{-3}$ \\
739: $^{232}_{90}\mathrm{Th}$ & 49.369 &19.400 & $2p_{1/2}$ & 0.416 & 6.93$\times 10^{-3}$\\
740: & &24.010 & $2p_{3/2}$ & 0.055 & 1.95$\times 10^{-3}$\\
741: \hline
742: & &12.405 & $2s_{1/2}$ & 0.033 & $7.99\times 10^{-3}$\\
743: $^{236}_{92}\mathrm{U}$ & 45.242 & 13.596 & $2p_{1/2}$ & 0.906 & 8.32$\times 10^{-3}$\\
744: & & 19.655 & $2p_{3/2}$ & 0.098 & 1.97$\times 10^{-3}$ \\
745: \hline
746: & & 12.073 & $2s_{1/2}$ & 0.039& $9.06\times 10^{-3}$\\
747: $^{238}_{92}\mathrm{U}$ & 44.916 & 13.262 & $2p_{1/2}$ & 1.055 & 9.35$\times 10^{-3}$\\
748: & &18.323 & $2p_{3/2}$ & 0.120 &2.32$\times 10^{-3}$ \\
749: \hline
750: & &6.888 & $2s_{1/2}$ & 0.147 & 1.79$\times 10^{-2}$ \\
751: $^{248}_{96}\mathrm{Cm}$ & 43.380 &8.190 & $2p_{1/2}$ & 2.936 & 1.55$\times 10^{-2}$ \\
752: & &14.203 & $2p_{3/2}$ & 0.240 & 2.94$\times 10^{-3}$ \\
753: \end{tabular}
754:
755: \end{ruledtabular}
756: \end{table*}
757:
758: We find that NRES resonance strengths for electron capture and
759: scattering on the $2p$ orbitals are up to two orders of magnitude larger
760: than the corresponding values for NEEC followed by $\gamma$ decay
761: of the nucleus for the highly charged actinides presented in
762: Table~\ref{U_LL}. The $2p$ orbitals of $^{232}_{90}\mathrm{Th}$,
763: $^{236}_{92}\mathrm{U}$, $^{238}_{92}\mathrm{U}$ and
764: $^{248}_{96}\mathrm{Cm}$ have a major role in the IC decay of the
765: excited nuclear state. The IC coefficients corresponding to the neutral
766: atom have values between $\alpha=327$ for the transition of
767: $^{232}_{90}\mathrm{Th}$ and $\alpha=984$ for the one of
768: $^{248}_{96}\mathrm{Cm}$. In few-electron configurations, however, the
769: strongly-bound inner-shell electrons have a more pronounced influence on
770: nuclear coupling to the atomic shells \cite{Philips}. One electron in
771: the $2p_{1/2}$ orbital of the highly charged ion of
772: $^{248}_{96}\mathrm{Cm}$, for instance, accounts already for a partial
773: IC coefficient of $\alpha=188$. Thus the two orders of magnitude
774: difference between $S_{\rm IC}$ and $S_{\gamma}$ in Table~\ref{U_LL} can
775: be traced back to the behaviour of the IC and radiative decay rates and
776: the IC coefficient. The relatively small energy of the nuclear
777: transitions leads to a low radiative decay rate. On the other hand, the
778: heavy actinides presented in Table~\ref{U_LL} are high-$Z$ nuclei with
779: large radial electronic integrals $R_{L,\kappa_d,\kappa}$ which lead to
780: significant IC rates. The largest NRES resonance strength is associated
781: with capture into the $2p_{1/2}$ orbital of $^{248}_{96}\mathrm{Cm}$,
782: with $S_{\rm IC}=2.94$~b~eV, which is on the same order of magnitude
783: with the corresponding values presented in Table~\ref{K} for $M1$
784: nuclear transitions.
785:
786: In the scenario considered so far, only few electrons fulfill the
787: resonance condition due to the very narrow natural width of the nuclear
788: excited state. A possibility to relax the resonance condition is given
789: by the capture of the electron into an excited bound state. We consider
790: therefore the more complicated three-step NRESX process depicted in
791: Figure~\ref{scheme2}, in which NEEC into an excited electronic state is
792: followed by $K\alpha$ x-ray emission and only subsequently by IC of
793: the captured electron. Since the capture and IC electronic states do not
794: coincide, the scattered free electron will have a different kinetic
795: energy than the incident electronic beam, with the difference in energy
796: being carried away by the x-ray photon. Furthermore, for the very heavy
797: actinide nuclei where the electron capture into the $K$ shell with the
798: excitation of the first collective excited level is not possible, NRESX
799: among $L$ subshells turns out to have several advantages for
800: experimental observation. The resonance strength for NRESX is calculated
801: using the expression in Eq.~(\ref{rs2}), where the required electronic
802: widths and x-ray transition rates are provided by the OSCL92 module of
803: the GRASP92 package.
804:
805: For the first group of isotopes in Table~\ref{K}, where capture into
806: and, consequently, IC from the $K$ shell is possible, we envisage NEEC
807: into the $L$ shell of bare ions. The x-ray decay of the captured
808: electron to the $K$ shell occurs orders of magnitude faster than the
809: nuclear de-excitation. In this case, the IC decay of the nucleus will
810: follow the x-ray emission and will ionize the bound electron from the
811: $K$ shell. In Tables \ref{L_E2} and \ref{L_M1} we present continuum
812: electron energies $E_c$, NEEC rates $Y_n$ and NRESX resonance strengths
813: $S$ for the capture into the $2s_{1/2}$, $2p_{1/2}$ and $2p_{3/2}$
814: orbitals of bare ions. Compared to the resonance strengths for the
815: two-step process of NRES presented in Table~\ref{K}, the values for
816: NRESX are several orders of magnitude smaller. The main reason for this
817: behaviour is the dependence of the resonance strength on the momentum of
818: the incoming continuum electron $p$, presented in Eq.~(\ref{tcs2}). All
819: isotopes in Tables \ref{L_E2} and \ref{L_M1} have nuclear excitation
820: energies that allow NEEC into the $K$ shell. Since the capture occurs
821: however into the $L$ shell, the continuum electron energy at the
822: resonance has large values, starting from the energy difference
823: between the $L$ and $K$ shells. As shown in the third column of Tables
824: \ref{L_E2} and \ref{L_M1}, the continuum electron energy has values
825: between approximately 45~keV for the case of $^{155}_{64}\mathrm{Gd}$
826: and going up to 114~keV for $^{187}_{75}\mathrm{Re}$.
827:
828: \begin{table*}[htb]
829: \caption{\label{L_E2}
830: Resonance strengths $S$ for NRESX: electron
831: recombination into the $L$ shell orbitals of bare ions followed by $K$-shell IC.
832: The nuclear transition multipolarity is $E2$. $E_{\rm exc}$ denotes the
833: nuclear exitation energy, $E_c$ is the continuum electron energy in
834: the center-of-mass frame and the capture orbital is denoted by $nl_j$.
835: In the fifth column we present the NEEC rate $Y_{n}$ and in the seventh the maximum
836: value of the convoluted cross section, $\tilde{\sigma}_{\rm max}$. The last column
837: contains total cross sections $\sigma_{bs}$ for the competing process of
838: bremsstrahlung.
839: }
840: \begin{ruledtabular}
841: \begin{tabular}{lrrclrrc}
842: Isotope & $E_{\rm exc}$(keV) & $E_{c}$(keV) & $nl_j$ & $Y_{n}$ (1/s) & $S$ (b eV) & $\tilde{\sigma}_{\rm max}$ (b) & $\sigma_{bs}$ (b) \\
843: \hline
844: & & 108.077 & $2s_{1/2}$ & $6.86\cdot 10^7$ & $8.44\times 10^{-4}$ & $3.36\times 10^{-5}$& \\
845: $^{154}_{64}\mathrm{Gd}$ & 123.071 & 108.063 & $2p_{1/2}$ & $1.14\cdot 10^8$ & $1.40\times 10^{-3}$ & $1.24\times 10^{-4}$& 12.7 \\
846: & & 108.946 & $2p_{3/2}$ & $1.51\cdot 10^8$ & $1.84\times 10^{-3}$ & $1.82\times 10^{-4}$& \\
847: \hline
848: & & 57.365 & $2s_{1/2}$ & $2.36\cdot 10^7$ & $1.08\times 10^{-3}$ & $4.30\times 10^{-5}$& \\
849: $^{164}_{66}\mathrm{Dy}$ & 73.392 & 57.348 & $2p_{1/2}$ & $1.85\cdot 10^8$ & $8.49\times 10^{-3}$ & $6.65\times 10^{-4}$& 13.5 \\
850: & & 58.357 & $2p_{3/2}$ & $2.54\cdot 10^8$ & $1.14\times 10^{-2}$ & $1.01\times 10^{-3}$& \\
851: \hline
852: & & 61.484 & $2s_{1/2}$ & $2.87\cdot 10^7$ & $1.11\times 10^{-3}$ & $4.42\times 10^{-5}$& \\
853: $^{170}_{68}\mathrm{Er}$ & 78.591 & 61.468 & $2p_{1/2}$ & $2.33\cdot 10^8$ & $9.04\times 10^{-3}$ & $6.26\times 10^{-4}$& 14.3 \\
854: & & 62.616 & $2p_{3/2}$ & $3.10\cdot 10^8$ & $1.18\times 10^{-2}$ & $9.30\times 10^{-4}$& \\
855: \hline
856: & & 58.241 & $2s_{1/2}$ & $9.88\cdot 10^5$ & $3.85\times 10^{-5}$ &$1.53\times 10^{-6}$ & \\
857: $^{174}_{70}\mathrm{Yb}$ & 76.471 & 58.222 & $2p_{1/2}$ & $1.10\cdot 10^7$ & $4.30\times 10^{-4}$ &$2.63\times 10^{-5}$ & 15.2 \\
858: & & 59.526 & $2p_{3/2}$ & $1.44\cdot 10^7$ & $5.48\times 10^{-4}$ &$3.91\times 10^{-5}$ & \\
859:
860: \hline
861: & & 73.780 & $2s_{1/2}$ & $3.60\cdot 10^7$ & $9.06\times 10^{-4}$ &$3.61\times 10^{-5}$ & \\
862: $^{178}_{72}\mathrm{Hf}$ & 93.180 & 73.759 & $2p_{1/2}$ & $2.76\cdot 10^8$ & $6.94\times 10^{-3}$ &$3.79\times 10^{-4}$ & 16.0 \\
863: & & 75.235 & $2p_{3/2}$ & $3.44\cdot 10^8$ & $8.47\times 10^{-3}$ &$5.43\times 10^{-4}$ & \\
864:
865: \hline
866: & & 82.939 & $2s_{1/2}$ & $4.22\cdot 10^7$ & $8.00\times 10^{-4}$ &$3.19\times 10^{-5}$ & \\
867: $^{180}_{74}\mathrm{W}$ & 103.557 & 82.915 & $2p_{1/2}$ & $2.95\cdot 10^8$ & $5.59\times 10^{-3}$ &$2.72\times 10^{-4}$ & 16.9 \\
868: & & 84.582 & $2p_{3/2}$ & $3.54\cdot 10^8$ & $6.56\times 10^{-3}$ &$3.85\times 10^{-4}$ & \\
869: \end{tabular}
870: \end{ruledtabular}
871: \end{table*}
872:
873: \begin{table*}[htb]
874: \caption{\label{L_M1}
875: Same as Table \ref{L_E2} for isotopes with $M1$ nuclear transitions.
876: }
877: \begin{ruledtabular}
878: \begin{tabular}{lrrcllcc}
879: Isotope & $E_{\rm exc}$(keV) & $E_{c}$(keV) & $nl_j$ & $Y_{n}(1/s)$ & $S$ (b eV) & $\tilde{\sigma}_{\rm max}$ (b) & $\sigma_{bs}$ (b)\\
880: \hline
881: & & 45.014 & $2s_{1/2}$ &$3.12\cdot 10^8$ & $2.61\times 10^{-2}$ &$1.04\times 10^{-3}$ & \\
882: $^{155}_{64}\mathrm{Gd}$ & 60.008 & 45.001 & $2p_{1/2}$ &$8.37\cdot 10^7$ & $7.01\times 10^{-3}$ &$6.22\times 10^{-4}$ & 12.7 \\
883: & & 45.883 & $2p_{3/2}$ &$8.39\cdot 10^7$ & $6.89\times 10^{-3}$ &$6.82\times 10^{-4}$ & \\
884:
885: \hline
886: & & 78.138 & $2s_{1/2}$ &$2.00\cdot 10^9$ & $6.75\times 10^{-2}$ &$2.69\times 10^{-3}$ & \\
887: $^{165}_{67}\mathrm{Ho}$ & 94.700 & 78.122 & $2p_{1/2}$ &$2.71\cdot 10^8$ & $9.17\times 10^{-3}$ &$6.73\times 10^{-4}$ & 13.9 \\
888: & & 79.198 & $2p_{3/2}$ &$1.47\cdot 10^8$ & $4.89\times 10^{-3}$ &$4.08\times 10^{-4}$ & \\
889:
890: \hline
891: & & 60.417 & $2s_{1/2}$ &$1.18\cdot 10^9$ & $6.85\times 10^{-2}$ &$2.73\times 10^{-3}$ & \\
892: $^{173}_{70}\mathrm{Yb}$ & 78.647 & 60.398 & $2p_{1/2}$ &$2.67\cdot 10^8$ & $1.54\times 10^{-2}$ &$9.40\times 10^{-4}$ & 15.2 \\
893: & & 61.702 & $2p_{3/2}$ &$2.17\cdot 10^8$ & $1.22\times 10^{-2}$ &$8.70\times 10^{-4}$ & \\
894:
895: \hline
896: & & 94.995 & $2s_{1/2}$ &$4.04\cdot 10^8$ & $9.72\times 10^{-3}$ &$3.87\times 10^{-4}$ & \\
897: $^{175}_{71}\mathrm{Lu}$ & 113.804 & 94.975 & $2p_{1/2}$ &$1.39\cdot 10^8$ & $3.35\times 10^{-3}$ &$1.94\times 10^{-4}$ & 15.6 \\
898: & & 96.363 & $2p_{3/2}$ &$1.32\cdot 10^8$ & $3.13\times 10^{-3}$ &$2.12\times 10^{-4}$ & \\
899:
900: \hline
901: & & 103.391 & $2s_{1/2}$ &$1.07\cdot 10^9$ & $2.26\times 10^{-2}$ &$9.00\times 10^{-4}$ & \\
902: $^{179}_{72}\mathrm{Hf}$ & 122.791 & 103.370 & $2p_{1/2}$ &$2.06\cdot 10^8$ & $4.34\times 10^{-3}$ &$2.37\times 10^{-4}$ & 16.0 \\
903: & & 104.846 & $2p_{3/2}$ &$1.39\cdot 10^8$ & $2.90\times 10^{-3}$ &$1.86\times 10^{-4}$ & \\
904:
905: \hline
906: & & 104.112 & $2s_{1/2}$ &$4.74\cdot 10^{9}$& $1.11\times 10^{-1}$ &$4.42\times 10^{-3}$ & \\
907: $^{185}_{75}\mathrm{Re}$ & 125.35 & 104.086 & $2p_{1/2}$ &$6.15\cdot 10^8$ & $1.43\times 10^{-2}$ &$6.60\times 10^{-4}$ & 17.4 \\
908: & & 105.857 & $2p_{3/2}$ &$2.22\cdot 10^8$ & $5.11\times 10^{-3}$ &$2.84\times 10^{-4}$ & \\
909: \hline
910:
911: & & 112.996 & $2s_{1/2}$ &$4.20\cdot 10^{9}$& $8.17\times 10^{-2}$ &$3.25\times 10^{-3}$ & \\
912: $^{187}_{75}\mathrm{Re}$ & 134.24 & 112.970 & $2p_{1/2}$ &$5.21\cdot 10^8$ & $1.01\times 10^{-2}$ &$4.60\times 10^{-4}$ & 17.4 \\
913: & & 114.741 & $2p_{3/2}$ &$1.68\cdot 10^8$ & $3.22\times 10^{-3}$ &$1.79\times 10^{-4}$ & \\
914:
915: \end{tabular}
916: \end{ruledtabular}
917: \end{table*}
918:
919: \begin{table*}[htb]
920: \caption{\label{U_ML} Resonance strengths $S$ for NRESX with capture
921: into the $L$- and $M$-shell $p$ orbitals of He-like ions followed by the
922: intra-shell or $L \alpha$ radiative decay to the $2s$ state and IC of
923: the bound electron. The nuclear transition multipolarity is $E2$.
924: $E_{\rm exc}$ denotes the nuclear exitation energy, $E_{c}$ is the
925: continuum electron energy in the center-of-mass frame and $nl_j$ stands
926: for the capture orbital.
927: %The last column contains bremsstrahlung total
928: %cross sections $\sigma_{bs}$.
929: }
930: \begin{ruledtabular}
931: \begin{tabular}{lrcccccl}
932: Isotope & $E_{\rm exc}$(keV) & $nl_j$ & $E_{c}$(keV) &$S$ (b eV) & $nl_j$ & $E_{c}$(keV) &$S$ (b eV) \\
933: \hline
934: $^{232}_{90}\mathrm{Th}$ &49.369 & $2p_{1/2}$ & 18.517 & 0.337 & $3p_{1/2}$ & 36.101 &0.059 \\
935:
936: $^{232}_{90}\mathrm{Th}$ & 49.369 & $2p_{3/2}$& 22.270 & 0.291 & $3p_{3/2}$ & 37.223 & 0.064 \\
937: \hline
938: $^{236}_{92}\mathrm{U}$ & 45.242 & $2p_{1/2}$& 12.688 & 0.894 & $3p_{1/2}$ & 31.270 & 0.125 \\
939:
940: $^{236}_{92}\mathrm{U}$ & 45.242 & $2p_{3/2}$& 16.872 & 0.683 & $3p_{3/2}$ & 32.519 & 0.131 \\
941: \hline
942: $^{238}_{92}\mathrm{U}$ & 44.916 & $2p_{1/2}$ & 12.362 & 1.049 & $3p_{1/2}$ & 30.941 & 0.144 \\
943:
944: $^{238}_{92}\mathrm{U}$ & 44.916 & $2p_{3/2}$& 16.540 & 0.797 & $3p_{3/2}$ & 32.190 & 0.151 \\
945: \hline
946: $^{248}_{96}\mathrm{Cm}$ & 43.380 & $2p_{1/2}$& 7.210 & 3.411 & $3p_{1/2}$ & 27.927 & 0.306 \\
947:
948: $^{248}_{96}\mathrm{Cm}$ & 43.380 & $2p_{3/2}$& 12.376 & 1.910 & $3p_{3/2}$ & 29.469 & 0.299 \\
949:
950: \end{tabular}
951:
952: \end{ruledtabular}
953: \end{table*}
954:
955: The case of the heavy actinide nuclei with low-lying first excited
956: states also offers other scenarios for NRESX. Since IC of the $K$-shell
957: electrons is energetically forbidden, we consider in the following
958: capture into initially He-like ions. A first scenario considers NEEC
959: into the $2p$ orbitals of He-like ions, followed by the fast x-ray decay
960: of the captured electron to the $2s_{1/2}$ state. The x-ray transition
961: for the considered highly charged ions occurs several orders of
962: magnitude faster than the nuclear decay. IC will therefore follow the
963: electronic transition and ionize the $2s_{1/2}$ electron. The initial
964: and final continuum electron energies are then given by the
965: corresponding energies of the IC and capture $L$ subshells. In a
966: similar manner, one can envisage the NEEC into the $3p$ orbitals of
967: He-like ions, with the subsequent decay of the captured electron to the
968: $2s$ ground state. In Table~\ref{U_ML} we present continuum electron
969: energies and NRESX resonance strengths for NEEC occurring into the $2p$
970: and $3p$ orbitals of He-like ${\rm Th}^{88+}$, ${\rm U}^{90+}$, and
971: ${\rm Cm}^{94+}$ ions. The resonance strengths for NEEC into the $2p$
972: orbitals are larger than the ones for capture into the $3p$ orbitals,
973: due to the smaller electron momentum values and different overlap of
974: the electronic wave functions with the nuclear matter. The largest
975: resonance strength value is the one for NRESX with capture into the
976: $2p_{1/2}$ orbital of the initially He-like ${\rm Cm}^{94+}$ ion,
977: namely, $S=3.41$~b eV.
978:
979: The initial and final states of NRESX coincide with those of
980: bremsstrahlung, where a photon is directly emitted in a continuum-continuum
981: transition. Bremsstrahlung is therefore a background process
982: which may complicate the observation of NRESX. In Tables \ref{L_E2} and \ref{L_M1}
983: we give total radiation cross section values calculated within a nonrelativistic
984: approximation (photon energy $\ll$ electron rest energy)~\cite{Koch&Motz} for
985: orientation. For comparison,
986: the maximum values of the NRESX cross sections convoluted with a 10~eV width Gaussian
987: electron energy distribution are given ($\tilde{\sigma}_{\rm max}$).
988: Bremsstrahlung cross sections are typically 4-6 orders of magnitude
989: larger than the NRESX cross sections at resonance.
990: However, since the lifetime of the NRESX process is dominated by long nuclear
991: mean-lives, it occurs on a much longer time scale than bremsstrahlung.
992: This fact may be exploited in a possible observation of nuclear-resonant electron scattering.
993:
994: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
995: % about experiment
996: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
997:
998: Regarding the possible experimental observation of NRES, the high
999: sensibility of electron spectrometry to strong magnetic fields restricts
1000: the choice of the experimental setup. The presence of magnetic fields
1001: in electron coolers of
1002: storage rings perturbs the electron trajectory and makes the detection
1003: of scattered electrons very difficult. At the storage ring facility at
1004: the GSI, electron scattering experiments have been
1005: performed using a gas target as electron target
1006: \cite{Hagmann_spectrometer,NIMB,Hagmann_prl}. The accelerated ions are cycling in
1007: the Experimental Storage Ring (ESR) with velocities close to the speed of
1008: light and are passing through a gas target with electron densities of
1009: $10^{12}-10^{14}$ electrons per cm$^2$. The quasifree electrons are then
1010: scattered by the highly charged ions. The main drawback related to the
1011: use of gas targets is the nuclear Coulomb excitation that occurs due
1012: to the target nuclei.
1013:
1014:
1015: For an envisaged NRES experiment, when the resonance energy condition
1016: is fulfilled, the quasifree electrons can be captured by the fast ion
1017: with the simultaneous excitation of the nucleus, being then carried away
1018: from the gas target. After a time interval corresponding to the mean
1019: lifetime of the nuclear excited state in the actual electronic configuration
1020: of the highly charged ion, the electron is expelled by the
1021: ion and can be detected using an electron spectrometer
1022: \cite{Hagmann_spectrometer}. The energy of the electron will be given by
1023: the transformation of the final continuum electron energy $E_c$ from the
1024: center-of-mass frame to the laboratory frame \cite{Eichler},
1025: %
1026: \begin{equation}
1027: E_c^{lab}/c=\gamma(E_c^{cm}/c+\beta
1028: p^{cm}\mathrm{cos}\theta) \, .
1029: \end{equation}
1030: %
1031: The emitted electron is
1032: characterized in the center-of-mass system by the momentum $p^{cm}$ with
1033: a direction determined by the polar angle $\theta$ with respect to the
1034: $z$ axis. In the equation above, $c$ stands for the speed of light and
1035: $\beta$ and $\gamma$ are the reduced velocity and the Lorentz factor of
1036: the ion, respectively. The electron spectrometer actually detects the
1037: electrons emitted in the forward direction \cite{Hagmann_prl,Hagmann_spectrometer}.
1038:
1039: The initial and final states of the scattering process in this scenario
1040: are the same as the ones of the non-resonant process of electron capture
1041: to the continuum (ECC)~\cite{ecc1,ecc2,ecc_doris}. This process is one
1042: of the major sources of background for NRES, and due to the identical
1043: initial and final states, quantum interference between the two processes
1044: may occur. However, an investigation of the corresponding time scales
1045: for the two processes reveals that in contrast to the nuclear
1046: lifetime-dependent NRES, ECC occurs much faster. Time-discrimination
1047: spectroscopy, as it has been proposed in Ref.~\cite{lifetime_prolongation}
1048: for NEEC followed by $\gamma$ decay of the nucleus, can therefore reduce
1049: substantially the ECC background. Similarly to the concept presented in Ref.
1050: \cite{lifetime_prolongation}, the different time scales of NRES and ECC
1051: have as a result a spatial separation of the electron emissions in a
1052: storage ring experiment. While the ECC photons will be emitted almost
1053: instantaneously in the region of the gas target, internal conversion
1054: will only occur later, after the ions have already travelled a certain
1055: distance in the ring. For the present electron spectrometer, where
1056: all forward-emitted electrons are detected approximately 90~cm after the gas target
1057: \cite{Hagmann_prl,Hagmann_spectrometer}, the separation of the signal and
1058: background events is challenging and requires a special extension of the experimental setup.
1059:
1060:
1061: Particularly interesting is the case of NRESX occcuring into the $2p$
1062: orbitals of He-like ions of heavy actinides, with resonance strengths
1063: presented in Table~\ref{U_ML}. The captured electron undergoes a fast
1064: x-ray decay to the $2s_{1/2}$ orbital (the decay rates
1065: %
1066: %calculated with the OSCL92 module of the GRASP92 package
1067: %
1068: are $1.95\times 10^{10}$
1069: s$^{-1}$ for the $2p_{1/2} \to 2s_{1/2}$ transition and $8.86\times
1070: 10^{14}$ s$^{-1}$ for the $2p_{3/2} \to 2s_{1/2}$ transition). The IC rate for the
1071: $2s$ orbital electron is much smaller than the one for $2p$ orbital
1072: electrons, so that the nuclear lifetime in the case of the $1s^2 2s$
1073: configuration is longer that the one for the $1s^2 2p$ capture
1074: configurations. The nuclear mean-lives of the $1s^2 2s$
1075: ion configuration of the four studied heavy actinides have values
1076: between $\tau$=13~ns for $^{248}_{96}\mathrm{Cm}$ and $\tau$=50~ns for
1077: $^{232}_{90}\mathrm{Th}$, corresponding to a spatial separation of
1078: approximately 13 to 50~cm.
1079:
1080: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1081:
1082: \section{\label{sum} Summary}
1083:
1084: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1085:
1086: We have considered nuclear-resonant electron scattering,
1087: i.e. nuclear excitation by electron capture followed by internal conversion,
1088: focusing on finding prospective isotopes
1089: for a possible experimental observation of the process. Theoretical total
1090: cross sections and resonance strengths for a number of capture scenarios
1091: and collision systems have been presented.
1092:
1093: In the first place, we have investigated the process of NRES involving
1094: $E2$ and $M1$ nuclear transitions with electron recombination into the
1095: electronic ground state. A comparison with resonance strengths for NEEC
1096: followed by the radiative decay of the nucleus shows that the two
1097: processes are typically on the same order of magnitude. For the specific
1098: cases of the heavy actinides studied, the IC nuclear decay channel
1099: prevails, and the NRES resonance strengths are between one and two
1100: orders orders or magnitude larger.
1101:
1102: A second scenario in which the electronic capture occurs into an excited
1103: electronic state and is followed by x-ray emission has also been
1104: investigated. Due to the large width of the excited electronic
1105: state, the continuum electron resonance energy condition is
1106: significantly relaxed. We have found that for the heavy actinide
1107: isotopes, NRESX with electronic capture and IC from different subshells of the $L$
1108: shell presents large resonance strength values. Furthermore, the
1109: possible experimental observation of NRES in storage rings, e.g. at
1110: the present and future ESR
1111: facility of the GSI Darmstadt has been discussed, devoting special attention to
1112: the electron target setup. A time-discrimination measurement at the
1113: ESR could be used for discerning the process of NRES from the background of other
1114: atomic physics processes, such as ECC. The most promising candidates for
1115: time-discrimination measurements were found to be the heavy actinide
1116: nuclei in a NRESX scenario involving the $L$-shell orbitals. While the
1117: calculated resonance strengths, on the order of 1~b\ eV, still make the
1118: observation of the NRES effect challenging, the advent of
1119: the new storage ring facility at GSI and the reported interest for
1120: electron spectroscopy experiments in the relativistic regime are strong
1121: arguments for the need of consistent theoretical predictions and
1122: experimental scenarios.
1123:
1124: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1125:
1126: \begin{acknowledgments}
1127:
1128: The authors would like to thank Alexander Voitkiv for helpful comments.
1129: A.P. is indebted to Stefan Schippers, Siegbert Hagmann and Christophor Kozhuharov for
1130: discussions concerning the experimental issues mentioned in this paper.
1131:
1132: \end{acknowledgments}
1133:
1134: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1135: %\appendix
1136: %\section{\label{A}}
1137: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1138: \bibliography{res}
1139:
1140: \end{document}
1141: