1: % Submission to Physical Review A
2: \documentclass[aps,pra,preprint,superscriptaddress,showpacs]{revtex4}
3: \usepackage{epsfig}
4: \usepackage{bm}
5: % \usepackage{showkeys}
6:
7: %--------------------------------------%
8: %
9: % Definitions
10: %
11: %--------------------------------------%
12:
13: \def\mpol#1{#1}
14: %
15: \def\sfc#1{ \textbf{... #1 ...} } % comment
16: \def\sfq#1{ \textbf{(?? #1 ??)} } % question
17: \def\sfi#1{ $\bullet\;$ \textit{#1}} % insertion
18: %
19: % -------------- our definitions ---------------- %
20: %
21:
22: \def\ket#1{ $ \left\vert #1 \right\rangle $ }
23: \def\ketm#1{ \left\vert #1 \right\rangle }
24: \def\bra#1{ $ \left\langle #1 \right\vert $ }
25: \def\bram#1{ \left\langle #1 \right\vert }
26: \def\spr#1#2{ $ \left\langle #1 \left\vert \right. #2 \right\rangle $ }
27: \def\sprm#1#2{ \left\langle #1 \left\vert \right. #2 \right\rangle }
28: \def\me#1#2#3{ $ \left\langle #1 \left\vert #2 \right\vert #3 \right\rangle $ }
29: \def\mem#1#2#3{ \left\langle #1 \left\vert #2 \right\vert #3 \right\rangle }
30: \def\rmem#1#2#3{ \left\langle #1 \left\vert \left\vert #2
31: \right\vert \right\vert #3 \right\rangle }
32: %
33: \def\threej#1#2#3#4#5#6{ $ \left( \matrix{ #1 & #2 & #3 \cr
34: #4 & #5 & #6 } \right) $ }
35: \def\threejm#1#2#3#4#5#6{ \left( \matrix{ #1 & #2 & #3 \cr
36: #4 & #5 & #6 } \right) }
37: %
38: \def\sixj#1#2#3#4#5#6{ $ \left\{ \matrix{ #1 & #2 & #3 \cr
39: #4 & #5 & #6 } \right\} $ }
40: \def\sixjm#1#2#3#4#5#6{ \left\{ \begin{array}{ccc}
41: #1 & #2 & #3 \\
42: #4 & #5 & #6
43: \end{array} \right\} }
44: %
45: \def\ninejm#1#2#3#4#5#6#7#8#9{ \left\{ \matrix{ #1 & #2 & #3 \cr
46: #4 & #5 & #6 \cr
47: #7 & #8 & #9 } \right\} }
48: %
49: \def\twobytwo#1#2#3#4{ \left( \begin{array}{cc}
50: #1 & #2 \\[0.2cm]
51: #3 & #4 \end{array} \right) }
52:
53: % \def\mpl#1{\textsc{#1}} % multipole
54: \def\mpl#1{\uppercase{#1}} % multipole
55: \def\etal{et al.} %
56:
57: \def\sfc#1{ \textbf{... #1 ...} } % comment
58: \def\sfq#1{ \textbf{(?? #1 ??)} } % question
59: \def\sfi#1{ $\bullet\;$ \textit{#1}} % insertion
60: \def\sfl#1{ {\large $\to\;$} \textsf{#1}} % logical sequence
61:
62: \def\asc#1{ { \textbf{... #1 ...} }} % my comments
63: \def\asi#1{ { $\bullet\;$ \textit{#1} }} % my insertion
64: \def\asr#1{ { \textbf{I would remove this: ... #1 ...} }} % to be removed
65:
66: %
67: \begin{document}
68:
69: %--------------------------------------%
70: %
71: % Title
72: %
73: %--------------------------------------%
74:
75: \title{Relativistic calculation of the
76: two-photon decay rate of highly-excited ionic states}
77:
78: %--------------------------------------%
79: %
80: % Authors
81: %
82: %--------------------------------------%
83:
84:
85: \author{Ulrich D. Jentschura}\email{ulj@mpi-hd.mpg.de}
86: \affiliation{Max-Planck-Institut f\"ur Kernphysik,
87: Postfach 103980, D-69029 Heidelberg, Germany}
88: \affiliation{Institut f\"{u}r Theoretische Physik, Philosophenweg 16, 69120 Heidelberg}
89: \author{Andrey Surzhykov}
90: \affiliation{Max-Planck-Institut f\"ur Kernphysik,
91: Postfach 103980, D-69029 Heidelberg, Germany}
92: \affiliation{Physikalisches Institut der Universit\"{a}t, Philosophenweg 12, 69120 Heidelberg}
93:
94: \date{\today}
95:
96: %--------------------------------------%
97: %
98: % Abstract
99: %
100: %--------------------------------------%
101:
102: \begin{abstract}
103: Based on quantum electrodynamics,
104: we reexamine the two-photon decay of one-electron atoms.
105: Special attention is paid to the calculation of the (two-photon)
106: total decay rates which can be viewed as the imaginary part of the
107: two-loop self-energy. We argue that our approach can easily be
108: applied to the cases with a virtual state having an
109: intermediate energy between the initial and the final state of the
110: decay process leading, thus, to the resonance peaks in the two-photon
111: energy distribution. In order to illustrate our approach, we obtain
112: fully relativistic results, resolved into electric and magnetic
113: multipole components, for the two-photon decay rates of the
114: $3S_{1/2} \to 1S_{1/2}$ transition in neutral hydrogen as well as
115: in various hydrogen-like ions.
116: \end{abstract}
117:
118:
119: \pacs{31.30.jn, 31.30.jc, 12.20.Ds, 32.80.Wr}
120:
121:
122: \maketitle
123: %--------------------------------------%
124: %
125: % Text
126: %
127: %--------------------------------------%
128:
129: \section{INTRODUCTION}
130:
131: Since the seminal work of
132: G\"oppert-Mayer \cite{GoM31}, two-photon
133: decay rates of excited states
134: in hydrogen-like atoms and ions have been the subject of intense
135: experimental \cite{LiN65, MaS72, PeD85, MoD04} and
136: theoretical \cite{BrT40,DrG81,FlE84,CrT86,FlS88,SaP98,Jen07,Jen07b}
137: studies. For many years, the investigations
138: have dealt not only with the total decay rates but also with the energy
139: and even angular distributions of the two emitted photons. By analyzing these
140: (two-photon) properties, unique information has been obtained about the
141: structural properties of one-electron systems including subtle relativistic
142: effects as well as about the basic concepts of quantum
143: physics such as, e.g., the entanglement.
144:
145: Even though large experimental and theoretical efforts have been
146: undertaken in the past to understand various aspects of the two-photon
147: decay of hydrogen-like atoms, the analysis of this process still raises a
148: number of unresolved problems. One of these problems, which currently attracts
149: a lot of interest, concerns those two-photon transition from highly excited
150: states to the ground state which pass through
151: an intermediate state with a lower energy than the
152: initial state of the two-photon process~\cite{CrT86,FlS88} and
153: can alternatively decay to the ground-state via two
154: (or more) sequential one-photon emissions.
155: Such a transition leads to resonance peaks in the energy spectrum
156: of the coherently emitted photons from the
157: two-photon decays which are located at the energies corresponding to
158: the (real) intermediate states. One of the most pronounced examples of such a
159: situation is the $3S_{1/2} \to 1S_{1/2}$ two-photon $E1E1$ transition for which
160: the differential (in energy) emission rate has singularities at energies
161: corresponding to the $3S_{1/2} \to 2P_{1/2} \to 1S_{1/2}$ and
162: $3S_{1/2} \to 2P_{3/2} \to 1S_{1/2}$ cascades. A proper treatment of these
163: singularities is obviously required for computing total decay rates obtained
164: after an integration over the energies of the coherently
165: emitted photons in the direct two-photon decay $3S_{1/2} \to 1S_{1/2}$.
166:
167: During the last two decades, the theoretical treatment of the resonances
168: in the energy distribution of the emitted photons has been discussed in a
169: number of places. In general, the decay rate $\Gamma_{i}$ of an initial
170: state $|i\rangle$ in a hydrogenlike atom is the sum
171: of a one-photon decay rate $\Gamma^{(1)}_{i}$ and a two-photon
172: contribution $\Gamma^{(2)}_{i}$,
173: %
174: %
175: \begin{equation}
176: \label{gamma_total}
177: \Gamma_{i} = \Gamma^{(1)}_{i} + \Gamma^{(2)}_{i} \, .
178: \end{equation}
179: %
180: %
181: The expression for $\Gamma^{(2)}_{i}$ as originally derived
182: in~\cite{GoM31} is easily seen to involve an integral over the
183: energies of the emitted photons, the sum of which has to be
184: equal to the energy difference of the initial and final states,
185: and a summation over all possible intermediate, virtual states.
186: In order to avoid problems with
187: non-integrable singularities, the authors of~\cite{CrT86}
188: have attributed $\Gamma^{(2)}_{i}$ only to the so-called non-resonant
189: intermediate transitions, in contrast to a summation over the
190: complete intermediate-state spectrum. The non-resonant transitions are given by
191: intermediate states of energy higher than the
192: energy $E_i$ of the initial state \cite{CrT86} (the ``resonant'' intermediate states,
193: which are involved in the one-photon cascade processes,
194: are explicitly excluded from the sum over
195: intermediate states). Based on this assumption, the non-resonant contribution
196: for the $3S_{1/2} \to 1S_{1/2}$ two-photon transition was
197: calculated as $\Gamma^{(2)}_{3S} = 8.2196 \, s^{-1}$ for hydrogen. Later, this
198: result has been also confirmed in the calculations by Florescu
199: and co-workers \cite{FlS88} who used a different method
200: for the summation over the ``non-resonant'' states.
201:
202: Although the results presented in Refs.~\cite{CrT86,FlS88}
203: are in mutual agreement,
204: they are both based on the interpretation of the
205: two-photon decay rate $\Gamma^{(2)}_{i}$ as a rate
206: generated only by non-resonant intermediate levels. In our manuscript,
207: we would like to propose an alternative way for the computation of the
208: (two-photon) total decay rates which leads to a
209: natural removal of the infinities otherwise introduced into the
210: expression for the two-photon decay rate. We apply here a fully
211: relativistic, quantum electrodynamical approach to re-investigate
212: the two-photon decay of highly-excited states of hydrogen-like atoms,
213: paying special attention to a careful handling of the
214: resonances infinitesimally displaced from Feynman's photon integration
215: contour (these singularities
216: exactly correspond to the problematic ``resonant'' intermediate states).
217: By making use of this approach, we obtain finite,
218: physically sensible results for the decay rates of the two-photon
219: $3S_{1/2} \to 1S_{1/2}$ transitions in neutral hydrogen as well as
220: in the various hydrogen-like ions. Apart of the leading, electric
221: dipole ($E1E1$) transition, we also discuss the contributions
222: from the higher multipole components to the total decay rate.
223:
224: This paper is organized as follows: after a brief survey of the
225: theoretical expressions used in our analysis (Sec.~\ref{theory}),
226: we proceed by discussing the method of evaluation (Sec.~\ref{evaluation})
227: as well as the numerical results obtained for the differential as
228: well as the total two-photon decay rates (Sec.~\ref{results}).
229: Conclusions are given in Sec.~\ref{conclusions}.
230:
231: %
232: % ------ here starts the theory part ---- %
233: %
234:
235: \section{THEORY}
236: \label{theory}
237:
238: Within quantum electrodynamics,
239: the (negative) imaginary part of the self-energy is just the $\Gamma/2$,
240: where $\Gamma$ is the decay width \cite{BaS78,BaS91,Jen04}.
241: The one-photon decay $\Gamma^{(1)}$ rate is obtained
242: from the imaginary part of the one-loop
243: self-energy, while the two-loop self-energy gives rise to the
244: two-photon decay rate $\Gamma^{(2)}$.
245: Because the relativistic formulation of the two-loop self-energy
246: problem has been discussed before in a number of places \cite{FoY73,YeI03},
247: we only mention here that by following a straightforward
248: generalization of the standard procedure
249: described for the non-relativistic framework in Refs.~\cite{Jen07, Jen07b,Jen04},
250: we obtain the following expression for the two-photon decay rate
251: ($\hbar = c = \epsilon_0 = 1$),
252: %
253: %
254: \begin{equation}
255: \label{gamma_2_QED}
256: \Gamma^{(2)}_i = \frac{\alpha^2}{\pi}
257: \lim\limits_{\epsilon \to 0} {\rm Re} \,
258: \int\limits_{0}^{\omega_{\rm max}}{\rm d}\omega_1 \,
259: \omega_1 \, \omega_2 \, \int {\rm d}\Omega_1 {\rm d}\Omega_2 \,
260: S_{if}(\omega_1, \omega_2) \, ,
261: \end{equation}
262: %
263: %
264: where $\omega_1 + \omega_2 = \omega_{\rm max} =
265: E_i - E_f$ with the initial and final state energies
266: $E_i$ and $E_f$, respectively. $S_{if}$ is given by
267: %
268: %
269: \begin{eqnarray}
270: \label{amplitude_general}
271: & & S_{if}(\omega_1, \omega_2) = \sum_{\nu}
272: \biggl( \frac{\mem{{\psi_{f}}}{\mathbf{A}^*_1}{\psi_\nu}
273: \mem{\psi_\nu}{\mathbf{A}^*_2}{\psi_{i}}}{E_{i} - E_{\nu}
274: - \omega_2 + i \epsilon}
275: \nonumber \\
276: & & \qquad + \frac{\mem{{\psi_{f}}}{\mathbf{A}^*_2}{\psi_\nu}
277: \mem{\psi_\nu}{\mathbf{A}^*_1}{\psi_{i}}}{E_{i} - E_{\nu}
278: - \omega_1 + i \epsilon} \biggr)
279: \nonumber \\
280: & & \times \sum_{\rho}
281: \biggl( \frac{\mem{{\psi_{i}}}{\mathbf{A}_1}{\psi_\rho}
282: \mem{\psi_\rho}{\mathbf{A}_2}{\psi_{f}}}{E_{i} - E_{\rho}
283: - \omega_1 + i \epsilon}
284: \nonumber \\
285: & & \qquad + \frac{\mem{{\psi_{i}}}{\mathbf{A}_2}{\psi_\rho}
286: \mem{\psi_\rho}{\mathbf{A}_1}{\psi_{f}}}{E_{i} - E_{\rho}
287: - \omega_2 + i \epsilon} \biggr) \, ,
288: \end{eqnarray}
289: %
290: %
291: where in the second factor, the initial and the final state
292: are exchanged, but the infinitesimal imaginary part in the
293: denominators remains $ + i \epsilon$ (i.e., does not change sign).
294: We here manifestly assume that
295: $\psi_{i}(\bm{r}) \equiv \psi_{n_i j_i \mu_i}(\bm{r})$ and
296: $\psi_{f}(\bm{r}) \equiv \psi_{n_f j_f \mu_f}(\bm{r})$ are the well-known
297: solutions of the Dirac Hamiltonian for a single electron in the
298: standard representation, describing an electron bound
299: to a point-like nucleus with
300: charge number $Z$. For photons propagating with wave vector $\bm{k}_i \;\, (i=1,2)$ and
301: unit polarization vector ${\bm u_{\lambda_i}}$
302: ($\bm{k}_i \cdot {\bm u}_{\lambda_i} = 0$), moreover, the electron-photon
303: interaction operator $\mathbf{A}_i$ in the transition amplitude
304: (\ref{amplitude_general}) can be written in velocity gauge as:
305: %
306: %
307: \begin{equation}
308: \label{interaction_operator}
309: \mathbf{A}_i = A_0 \, \bm{\alpha} \cdot \bm{u}_{\lambda_i}
310: {\rm e}^{i \bm{k}_i \bm{r}} \, ,
311: \end{equation}
312: %
313: %
314: where $A_0$ is a normalization factor, ${\bm \alpha}$ are the
315: standard Dirac matrices, and $\lambda_i = \pm 1$ denotes
316: the \textit{helicity}, i.e.\ the
317: spin projection of the photon onto the direction of propagation $\bm{k}_i$.
318: It is important to note that even though the electron-photon
319: interaction operator (\ref{interaction_operator}) depends, of course,
320: on the direction of the photon emission, one has to integrate
321: over these directions in Eq.~(\ref{gamma_2_QED}) in order to get
322: the total decay rate.
323:
324: \section{EVALUATION}
325: \label{evaluation}
326:
327: The summation
328: over the intermediate states in the amplitude (\ref{amplitude_general})
329: runs over the \textit{complete} one-particle spectrum
330: $\ketm{\psi_{\nu}} \equiv \ketm{\psi_{n_\nu j_\nu \mu_\nu}}$, including a
331: summation over the discrete part of the spectrum as well as an integration
332: over the positive and negative-energy
333: continuum of the Dirac spectrum. One has to use the
334: full Dirac--Coulomb Green function---which is not known
335: in closed analytic form---in order to perform
336: this calculation consistently.
337: In recent years, the Green's function method~\cite{DrS91} has been widely
338: applied for the analysis of the total two-photon decay rates as
339: well as the photon-photon angular correlation functions \cite{SuK05}.
340: Various possibilities for the numerical implementation of the
341: relativistic Green's function are known, among which we would
342: like to mentionas
343: (i)~a well-known formulation in terms of Whittaker
344: functions~\cite{PJM74} and (ii) a Sturmian decomposition in
345: terms of Laguerre polynomials as suggested by Hylton and
346: Snyderman~\cite{HyS97}.
347:
348:
349: %
350: %
351: \begin{figure}[t]
352: \begin{center}
353: \includegraphics[width=1.0\columnwidth, angle=0]{Fig1.eps}
354: \end{center}
355: \vspace*{-0.5cm} \caption{(Color online) Differential decay rate
356: ${\rm d}\Gamma/{\rm d}x$ for the dominant E1E1 component of the
357: $3S \to 1S$ two-photon
358: transitions in neutral hydrogen as well as in hydrogen-like
359: ions with $Z = 1, 10, 20, 30, 35$ and $40$, where $x$ is the
360: energy sharing $x = \omega_1/(\omega_1+\omega_2)$.
361: Relativistic wave functions are used for the initial, intermediate
362: and final states, but the electron-photon interaction has been restricted to
363: electric dipole term ($E1 E1$ term). The resonance peaks in the decay rate
364: correspond to the resonant $3S_{1/2} \to 2P_{1/2} \to 1S_{1/2}$ and
365: $3S_{1/2} \to 2P_{3/2} \to 1S_{1/2}$ decay processes.}
366: \end{figure}
367: %
368: %
369:
370: We use the latter representation and apply the techniques of
371: Racah's algebra to all
372: spherical tensors and to the standard radial-angular representation of the wave
373: functions, and to the Dirac--Coulomb Green function.
374: For the interaction of electrons with the radiation
375: field, the spherical tensor components are obtained from the known standard
376: multipole expansion of the photon operator [see, e.g., Eq. (5) of
377: Ref.~\cite{eic98}],
378: %
379: %
380: \begin{eqnarray}
381: \label{photon_operator_decomposition}
382: \bm{u}_{\lambda}{\rm e}^{i \bm{k} \bm{r}} =
383: \sqrt{2 \pi} \sum\limits_{L M p} i^L (i \lambda)^p
384: \sqrt{2L + 1}
385: \bm{A}_{LM}^{(p)} \, D^{L}_{M \lambda}(\bm{n}) \, ,
386: \end{eqnarray}
387: %
388: %
389: where $\bm{A}_{LM}^{(p)}$ denotes the electric ($p$ = 1) and magnetic ($p$ = 0)
390: multipole fields, respectively.
391:
392: %
393: % Results
394: %
395: \section{RESULTS}
396: \label{results}
397:
398: The great advantage of the multipole decomposition
399: (\ref{photon_operator_decomposition}) is that is allows us
400: to study the contributions to the total (two-photon) decay rate
401: from the various \textit{allowed} multipole combinations. We use
402: the integrand in the integral over $\omega_1$ in
403: Eq.~(\ref{amplitude_general}) as a measure of the differential decay
404: rate (where we can set explicitly $\epsilon = 0$ for the differential
405: rate). The energy distributions of the two photons emitted in the
406: $3S_{1/2} \to 1S_{1/2}$ decay of neutral hydrogen and hydrogen-like
407: ions are calculated as a function of the energy
408: sharing parameter $x = {\omega_1}/({\omega_1}+{\omega_2})$.
409: For an energy sharing in the range $0 < x < 1$, the contributions to
410: the energy distribution from the $E1 E1$ and $E1 M2$ multipole combinations
411: are displayed in Figs.~1 and 2, respectively.
412: As seen from these figures, the photon energy distributions
413: for both multipole combinations exhibit sharp resonance peaks.
414: As already mentioned, this behaviour is due to the fact that the summation in
415: Eq.~(\ref{amplitude_general}) includes also intermediate
416: states $\ketm{\nu}$ having an energy $E_{\nu}$ with $E_i > E_{\nu} >E_f$.
417: However, the intermediate states contributing to the peaks
418: are not only defined by the (symmetry of)
419: the initial $\ketm{i}$ and final $\ketm{f}$ states but also by
420: the multipole components of the radiation field involved in the
421: two-photon process and are different for
422: $E1 E1$ as opposed to $E1 M2$, and in addition,
423: marked differences exist between the low-$Z$ and the high-$Z$ region. E.g.,
424: the fine-structure of the
425: resonance in the $E1 E1$ energy spectrum grows with $Z$
426: (the contributing states are $2P_{1/2}$ and $2P_{3/2}$).
427: By contrast, no splitting is observed---even for very heavy ions---for the
428: $E1 M2$ component of the $3S_{1/2} \to 1S_{1/2}$ decay.
429: Only one intermediate state, namely $2P_{3/2}$, is allowed for $E1 M2$.
430:
431: %
432: %
433: \begin{figure}[t]
434: \begin{center}
435: \includegraphics[width=1.0\columnwidth, angle=0]{Fig2.eps}
436: \end{center}
437: \vspace*{-0.5cm} \caption{(Color online)
438: Differential decay rate ${\rm d}\Gamma/{\rm d}x$
439: for the E1M2 component of the $3S \to 1S$
440: two-photon transitions in neutral hydrogen as well as in hydrogen-like
441: ions with $Z = 1, 10, 20, 30, 35$ and $40$
442: ($x$ is the fraction of energy carried by the E1 photon).
443: The $E1 M2$ term is treated relativistically. The resonances in the decay rate
444: exclusively correspond to the $3S_{1/2} \to 2P_{3/2} \to 1S_{1/2}$ cascade, the
445: $2P_{1/2}$ state does not contribute.}
446: \end{figure}
447: %
448: %
449:
450: Our treatment of the intermediate state resonance peaks
451: in the integration over the photon energy
452: is dictated by an accurate analysis of
453: Eqs.~(\ref{gamma_2_QED})--(\ref{amplitude_general})
454: as obtained form the relativistic two-loop self energy.
455: The general structure of the contribution to $
456: S_{if}(\omega_1, \omega_2)$ generated by
457: virtual states with intermediate energies $E_i > E_{\nu} > E_f$
458: with resonance energy $\omega_r = E_i - E_{\nu}$ or
459: $\omega_r = E_{\nu} - E_f$ is as follows:
460: %
461: %
462: \begin{eqnarray}
463: \label{resonant_terms}
464: S_{if}(\omega_1, \omega_2) &\sim&
465: \frac{R_1}{\omega_r - \omega + i \epsilon}
466: + \frac{R_2}{\left(\omega_r - \omega + i \epsilon \right)^2} \, .
467: \end{eqnarray}
468: %
469: %
470: The integration of the first term can be carried out
471: using the Dirac prescription (see, for example,
472: Ref.~\cite{Jen07}). The second term of Eq.~(\ref{resonant_terms}) can
473: be treated using the formula
474: %
475: %
476: \begin{equation}
477: \label{quadratic_term_integration}
478: \lim\limits_{\epsilon \to 0} {\rm Re} \,
479: \int\limits_{0}^{1} {\rm d}\omega
480: \left(\frac{1}{\omega_r - \omega + i \epsilon} \right)^2 =
481: \frac{1}{\omega_r (\omega_r - 1)} \, ,
482: \end{equation}
483: %
484: %
485: where we used an appropriate scaling of the photon energy integration
486: variable in order to map the integration region
487: to the interval $(0,1)$.
488: It is important to note that Eq.~(\ref{quadratic_term_integration})
489: holds strictly for $0 < \omega_r < 1$, but the limit is
490: not approached uniformly~\cite{Jen07,Jen07b}; i.e.,~it
491: would be forbidden to exchange
492: the sequence of the limit $\epsilon \to 0$ with the
493: integration over $\omega$. As usual in quantum electrodynamic processes,
494: all regulators have to be kept
495: up until the very end of the calculation.
496:
497: With these preparations, it is easy now to integrate
498: over the photon energies (see Table 1
499: for the $3S_{1/2} \to 1S_{1/2}$ process).
500: As seen from the table, the cross sections for the
501: $E1 E1$, $E1 M2$ and $M1 M1$ components
502: components scale with the nuclear charge as $Z^6$, $Z^{10}$
503: and $Z^{10}$, respectively. As expected, this scaling behaviour
504: is similar to the $Z$-scaling of the multipole components
505: in the $2S_{1/2} \to 1S_{1/2}$ transition.
506:
507: %
508: %
509: \begin{table}[t]
510: \vspace*{1cm}
511: \begin{tabular}{c@{\hspace*{0.5cm}}c@{\hspace*{0.5cm}}c@{\hspace*{0.5cm}}c}
512: \hline
513: \hline
514: $Z$ & $E1E1$ & $E1M2$ & $M1M1$ \\
515: & (Z$^{-6}$) & (Z$^{-10}$ 10$^{10}$) & (Z$^{-10}$ 10$^{12}$) \\
516: \hline
517: 1 & 2.08 & 1.19 & 6.13 \\
518: 5 & 2.03 & 1.18 & 6.13 \\
519: 10 & 1.98 & 1.16 & 6.14 \\
520: 15 & 1.94 & 1.14 & 6.16 \\
521: 20 & 1.90 & 1.12 & 6.20 \\
522: 25 & 1.84 & 1.08 & 6.24 \\
523: 30 & 1.79 & 1.03 & 6.30 \\
524: 35 & 1.67 & 0.96 & 6.39 \\
525: 40 & 1.60 & 0.86 & 6.50 \\
526: \hline
527: \hline
528: \end{tabular}
529: \caption{\label{table1} Contributions from different combinations of multipoles
530: to the integrated decay rate $\Gamma^{(2)}$, in units of radians per second.
531: Relativistic calculations have
532: been performed for different hydrogen-like ions.}
533: \end{table}
534: %
535: %
536:
537: Furthermore, as seen from Table 1 and as implied by the
538: non-uniform convergence of the integrals,
539: the intermediate states with the energies $E_\nu$ lying between
540: the energies of the initial and the final states give a
541: \textit{finite} contribution to the two-photon decay rate.
542: For the electric dipole ($E1 E1$) transition in a neutral
543: hydrogen atom, e.g., a proper treatment of the
544: intermediate $2P_{1/2}$ and $2P_{3/2}$ states leads to the
545: decay rate of $\Gamma^{(2)}_{3S} = 2.08 \, s^{-1}$ which is in
546: agreement with the result of nonrelativistic
547: calculations reported in Ref.~\cite{Jen07}. However, when comparing
548: our prediction with the theoretical data by Cresser and co-workers \cite{CrT86}
549: a large discrepancy by about a factor of $4$ is observed.
550: The occurrence of the discrepancy is natural because the problematic virtual states
551: with intermediate energies are treated differently in \cite{CrT86}.
552:
553: %
554: % ------- conclusion --------- %
555: %
556:
557: \section{CONCLUSIONS}
558: \label{conclusions}
559:
560: The two-photon decay of hydrogen-like ions
561: has been re-investigated within the framework of relativistic
562: quantum electrodynamics. Starting from first principles of
563: this theory, we treat the total (two-photon) decay rate as the
564: imaginary part of the relativistic two-loop self-energy.
565: The great advantage of this approach, which has its
566: roots in field theory, is that it provides a
567: simple and efficient route to handle the potentially problematic
568: cases of those two-photon transitions from an excited
569: into the ground state which pass intermediate states that
570: can otherwise also be reached in one-photon cascades from the
571: initial to the final states.
572: We found that those states with energies that lie between the
573: energy of the initial and the final states, contribute a finite
574: correction to the total two-photon decay rate. Taking into
575: account this correction, we calculate the rates for
576: the $3S_{1/2} \to 1S_{1/2}$ two-photon decay of neutral hydrogen
577: as well as hydrogen-like ions.
578: Our results are in a good agreement with nonrelativistic
579: calculations for low $Z$ (see Refs.~\cite{Jen07,Jen07b})
580: but show a significant
581: deviation from the data by Cresser and co-workers
582: \cite{CrT86}.
583:
584: Our quantum electrodynamics approach, as discussed in the
585: present paper, opens a way for a \textit{systematic} theoretical
586: analysis of the simultaneous, coherent two-photon emission
587: from one-electron (and many-electron) atomic systems,
588: even in cases where problematic intermediate states with an energy
589: between the initial and final states give rise to resonance
590: peaks in the photon energy distributions.
591: We stress here that a conceivable alternative approach
592: to the removal of the formal infinities generated by the
593: intermediate ``cascade'' states, which is based on the
594: explicit removal of these states from the sums over $\nu$ and $\rho$
595: in Eq.~(\ref{gamma_2_QED}), gives rise to a number of conceptual
596: problems, including gauge-noninvariance with respect to length and
597: velocity gauges~\cite{Jen07b}. Our approach is manifestly gauge invariant
598: and also avoids problems connected with the identification of the infinitesimal
599: parts $i\epsilon$ in the propagator denominators in Eq.~(\ref{gamma_2_QED})
600: with partial or total decay rates of the intermediate states:
601: the $\epsilon$ parameters are free parameters which approach
602: zero after all other operations, including the integrations
603: over the photon energies, have been performed. This operation
604: leads to a finite result and corresponds, as explained in
605: Ref.~\cite{Jen07b}, to a {\em partial} removal of the
606: problematic intermediate states from the sum over all virtual
607: states involved in the two-photon process, albeit in a fully
608: gauge-invariant manner.
609:
610: In addition to its relevance for atomic physics, our approach may have
611: a significant impact for astrophysical studies where a detailed
612: knowledge of the (properties of) two-photon transitions is highly
613: required for the analysis of cosmological hydrogen and helium
614: recombination. The contribution of two-photon processes to
615: the recombination history represents an issue which
616: has recently attracted substantial theoretical
617: interest \cite{DuG05}.
618:
619:
620: \section*{ACKNOWLEDGMENTS}
621:
622: U.D.J. acknowledges support from Deutsche Forschungsgemeinschaft (Heisenberg
623: program), and A.S. acknowledges support from the Helmholtz Gemeinschaft
624: (Nachwuchsgruppe VH--NG--421). The authors acknowledge insightful discussions
625: with Z. Harman regarding the Sturmian decomposition of the Dirac--Coulomb Green
626: function as given in Ref.~\cite{HyS97}, and helpful discussions with P.
627: Indelicato on general aspects of the two-photon decay and associated
628: resonances.
629:
630: %
631: %
632: %
633: %
634: %
635: %
636: \begin{thebibliography}{10}
637:
638: \bibitem{GoM31} M.~G\"oppert-Mayer,
639: Ann. Phys. (Leipzig) {\bf 401}, 273 (1931).
640:
641: \bibitem{LiN65} M.~Lipeles, R.~Novic, and N.~Tolk,
642: Phys. Rev. Lett. \textbf{15}, 690 (1965).
643:
644: \bibitem{MaS72} R.~Marrus and R.~W.~Schmieder,
645: Phys. Rev. A {\bf 5}, 1160 (1972).
646:
647: \bibitem{PeD85} W.~Perrie, A.~J.~Duncan, H.~J.~Beyer,
648: and H.~Kleinpoppen,
649: Phys. Rev. Lett. {\bf 54}, 1790 (1985).
650:
651: \bibitem{MoD04} P.~H.~Mokler and R.~W.~Dunford,
652: Phys. Scr. {\bf 69}, C1 (2004).
653:
654: \bibitem{BrT40} G.~Breit and E.~Teller,
655: Astrophys. J. {\bf 91}, 215 (1940).
656:
657: \bibitem{DrG81} G.~W.~F.~Drake and S.~P.~Goldman,
658: Phys.\ Rev.\ A {\bf 23}, 2093 (1981).
659:
660: \bibitem{FlE84} V. Florescu, Phys. Rev. A {\bf 30}, 2441 (1984);
661: J.~H. Tung, X.~M. Ye, G.~J. Salamo, and F.~T. Chan,
662: Phys. Rev. A {\bf 30}, 1175 (1984);
663: V. Florescu, S. Patrascu, and O. Stoican,
664: Phys. Rev. A {\bf 36}, 2155 (1987).
665:
666: \bibitem{CrT86} J.~D.~Cresser, A.~Z.~Tang, G.~J.~Salamo, and F.~T.~Chan,
667: Phys. Rev. A {\bf 33}, 1677 (1986).
668:
669: \bibitem{FlS88} V.~Florescu, I.~Schneider, and I.~N.~Mihailescu,
670: Phys. Rev. A \textbf{38}, 2189 (1988).
671:
672: \bibitem{SaP98} J.~P.~Santos, F.~Parente, and P.~Indelicato,
673: Eur. Phys. J. D \textbf{3}, 43 (1998).
674:
675: \bibitem{Jen07} U.~D.~Jentschura, J. Phys. A {\bf 40}, F223 (2007).
676:
677: \bibitem{Jen07b} U.~D.~Jentschura, J. Phys. A {\bf 41}, 155307 (2008).
678:
679: \bibitem{BaS78} R.~Barbieri and J.~Sucher,
680: Nucl. Phys. B \textbf{134}, 155 (1978).
681:
682: \bibitem{BaS91} A.~O.~Barut and Y.~I.~Salamin,
683: Phys. Rev. A \textbf{43}, 2524 (1991).
684:
685: \bibitem{Jen04} U.~D.~Jentschura,
686: Phys. Rev. A \textbf{69}, 052118 (2004).
687:
688: \bibitem{FoY73} J.~A.~Fox and D.~R.~Yennie,
689: Ann. Phys. \textbf{81}, 438 (1973).
690:
691: \bibitem{YeI03} V.~A.~Yerokhin, P.~Indelicato, and V.~M.~Shabaev,
692: Eur. Phys. J. D \textbf{25}, 203 (2003).
693:
694: \bibitem{DrS91} R. A. Swainson and G. W. F. Drake,
695: J. Phys. A {\bf 24}, 95 (1991).
696:
697: \bibitem{SuK05} A.~Surzhykov, P.~Koval, and S.~Fritzsche,
698: Phys. Rev A \textbf{71}, 022509 (2005).
699:
700: %\bibitem{GoD81} S.~P.~Goldman and G.~W.~Drake,
701: % Phys. Rev. A \textbf{24}, 183 (1981).
702:
703: \bibitem{PJM74} P. J. Mohr, Ann. Phys. (N.Y.) {\bf 88}, 26 (1974);
704: Ann. Phys. (N.Y.) {\bf 88}, 52 (1974).
705:
706: \bibitem{HyS97} D.~J.~Hylton and N.~J.~Snyderman,
707: Phys. Rev. A \textbf{55}, 2651 (1997).
708:
709: \bibitem{eic98} J. Eichler, A. Ichihara, and T. Shirai,
710: Phys. Rev. A {\bf 58}, 2128 (1998).
711:
712: \bibitem{DuG05} V.~K.~Dubrovich and S.~I.~Grachev,
713: Astron. Lett. \textbf{31}, 359 (2005);
714: W.~Y.~Wong and D.~Scott, Mon. Not. R. Astron. Soc.
715: \textbf{375}, 1441 (2007);
716: J. Chluba and R.~A. Sunyaev,
717: e-print 0705.3033 [astro-ph].
718:
719: \end{thebibliography}
720:
721:
722: \end{document}
723: