physics0101044/part1
1: \documentstyle[preprint,pra,aps]{revtex}
2: \begin{document}
3: \draft
4: \title{\bf Calculations of parity nonconserving $s$-$d$ transitions 
5: in Cs, Fr, Ba$^+$, and Ra$^+$}
6: \author{V.A. Dzuba, V.V. Flambaum, and J.S.M. Ginges}
7: \address{School of Physics, University of New South Wales, 
8: Sydney 2052,Australia}
9: \date{\today}
10: \maketitle
11: 
12: \tightenlines
13: 
14: \begin{abstract}
15: 
16: We have performed {\it ab initio} mixed-states and sum-over-states 
17: calculations of parity nonconserving (PNC) electric dipole (E1) 
18: transition amplitudes between $s$-$d$ electron states of Cs, Fr, 
19: Ba$^{+}$, and Ra$^{+}$. 
20: For the lower states of these atoms we have also calculated 
21: energies, E1 transition amplitudes, and lifetimes.
22: We have shown that PNC E1 transition amplitudes between $s$-$d$ 
23: states can be calculated to high accuracy.
24: Contrary to the Cs $6s$-$7s$ transition, in these transitions there are no 
25: strong cancelations between different terms in the sum-over-states approach. 
26: In fact, there is one dominating term which deviates from the sum by less 
27: than $20\%$. This term corresponds to an $s$-$p_{1/2}$ weak matrix 
28: element, which can be calculated to better than $1\%$, 
29: and a $p_{1/2}$-$d_{3/2}$ E1 transition amplitude, which can be measured.
30: Also, the $s$-$d$ amplitudes are about four times larger than the 
31: corresponding $s$-$s$ transitions.
32: We have shown that by using a hybrid mixed-states/sum-over-states 
33: approach the accuracy of the calculations of PNC $s$-$d$ amplitudes 
34: could compete with that of Cs $6s$-$7s$ if $p_{1/2}$-$d_{3/2}$ 
35: E1 amplitudes are measured to high accuracy.
36: 
37: \end{abstract} 
38: \vspace{1cm}
39: \pacs{PACS: 32.80.Ys,31.15.Ar}
40: %*************************************************************************  
41: 
42: \section{Introduction}
43: 
44: Precise low-energy experiments on parity nonconservation (PNC) in atoms 
45: provide a test of the standard model of elementary particle interactions. 
46: By measuring PNC electric dipole (E1) transition amplitudes, the value 
47: of the nuclear weak charge can be extracted by comparison with 
48: calculations. In a recent PNC experiment with cesium \cite{wood} the 
49: PNC E1 transition amplitude between the $6s$ and $7s$ states has been 
50: determined with an unprecedented accuracy of 0.35\%. However, 
51: interpretation of the experiment is limited by 
52: the accuracy of the atomic calculations. Since 1989, calculations 
53: of the $6s$-$7s$ transition in Cs have been at the 1\% level 
54: \cite{flam89pnc,johns90}. At this level of accuracy the value of the 
55: nuclear weak charge is consistent with that predicted by the Standard Model. 
56: Recent measurements of values relevant to the PNC E1 amplitude 
57: (E1 transition amplitudes, hyperfine structure constants) 
58: are in much better agreement with the calculated values than they were 
59: ten years ago. From this it has been claimed that the accuracy of the 
60: calculated PNC E1 amplitude is 0.4\% \cite{bw}. Re-interpreting the Cs
61: measurement with the higher accuracy, while using the calculations 
62: \cite{flam89pnc,johns90}, the value of the nuclear weak charge gives a 
63: $2.5~\sigma$ deviation from the Standard Model prediction \cite{bw}. 
64: However, inclusion of the Breit interaction into the calculations 
65: reduces the deviation by about $1~\sigma$ \cite{Der,DzJohns}.  
66: Note that these measurements give the best limits on new physics beyond 
67: the Standard Model, such as extra Z bosons, leptoquarks, composite fermions 
68: \cite{casal,rosner,erler}.
69: 
70: One obviously needs an independent confirmation of the Cs result. 
71: In this paper we show that the accuracy of calculations of PNC E1 
72: transitions between $s$-$d$ states of 
73: Cs, Fr, Ba$^{+}$, and Ra$^{+}$ could compete 
74: with that of the Cs $6s$-$7s$ transition. 
75: The experiment for the $6s$-$5d$ transition in Ba$^+$ is 
76: already in progress \cite{fortson}.
77: 
78: \section{Many-body calculations}
79: 
80: We perform calculations for $N$-electron atoms with one external electron 
81: above closed shells.
82: The calculations start from the relativistic Hartree-Fock (RHF) method 
83: in the $\hat{V}^{N-1}$ approximation. The single-electron RHF Hamiltonian is 
84: \begin{equation}
85: \label{eq:RHF}
86: \hat{H}_{0}=c{\bf \alpha}\cdot \hat{{\bf p}}+(\beta -1)c^{2}-
87: Z/r+\hat{V}^{N-1},
88: \end{equation} 
89: ${\bf \alpha}$ and $\beta$ are Dirac matrices and $\hat{{\bf p}}$ is the 
90: electron momentum.
91: The accuracy of RHF energies is of the order of $10\%$ for heavy atoms 
92: like Cs, Fr, Ba$^+$, and Ra$^+$. 
93: 
94: In order to obtain more realistic wavefunctions, 
95: electron-electron correlations must be taken into account.
96: Correlation corrections to the electron orbitals are calculated 
97: using the ``correlation potential'' method 
98: \cite{flam87}. 
99: This method corresponds to adding a non-local correlation potential 
100: $\hat{\Sigma}$ to the potential $\hat{V}^{N-1}$ in the RHF equation
101: (\ref{eq:RHF}) and then solving for the states of the external electron. 
102: The correlation potential is defined such that its 
103: average value coincides with the correlation correction to energy, 
104: $\delta E_{i}=\langle\psi _{i}|\hat{\Sigma}|\psi _{i}\rangle$.
105: The correlation potential is calculated by means of many-body 
106: perturbation theory in the residual Coulomb interaction
107: \begin{equation}
108: U=\hat{H}-\sum_{i=1}^{N}\hat{H}_{0}({\bf r}_{i})=
109: \sum _{i<j}^{N}\frac{1}{|{\bf r}_{i}-{\bf r}_{j}|}-
110: \sum _{i=1}^{N}\hat{V}^{N-1}({\bf r}_{i}),
111: \end{equation}
112: where $\hat{H}$ is the exact Hamiltonian of an atom.
113: The lowest-order correlation diagrams (second-order in $U$) 
114: are presented in Fig.~\ref{fig:2ndorder}. At this level of calculation 
115: the accuracy for energy levels is about 1\%.
116: 
117: Using the correlation potential method and the Feynman diagram technique 
118: we include three series of higher order diagrams which are calculated 
119: in all orders of perturbation theory 
120: \cite{flam89pnc,flam89energy,flam89e1hfs}. 
121: These are screening of the electron-electron interaction, the hole-particle 
122: interaction, and chaining of the self-energy operator $\hat{\Sigma}$. 
123: The electron-electron screening (see Fig.~\ref{fig:screening}) and the 
124: hole-particle interaction (Fig.~\ref{fig:hpchain}) are incorporated into 
125: the self-energy operator $\hat{\Sigma}$ (Fig.~\ref{fig:hpscreense}). 
126: Chaining of the self-energy operator to all orders 
127: (Fig.~\ref{fig:sechain}) is then calculated by adding 
128: $\hat{\Sigma}$ to the Hartree-Fock potential $\hat{V}^{N-1}$ 
129: and solving the equation
130: \begin{equation}
131: (\hat{H}_{0}+\hat{\Sigma}-\epsilon)\psi =0
132: \end{equation}
133: iteratively for the states of the external electron.
134: In this way ``Brueckner'' energies and orbitals are obtained. 
135: These energies have an accuracy of the order of $0.1\%$.
136: The wavefunctions can be further modified by placing a coefficient before 
137: $\hat{\Sigma}$ such that the corresponding energy coincides with the 
138: experimental value. This fitting of the Brueckner orbitals can be considered 
139: as a way of including higher-order diagrams into the calculations. 
140: 
141: We use the time-dependent Hartree-Fock method (which is equivalent to the 
142: random-phase approximation with exchange) to calculate the interaction of 
143: external fields with atomic electrons.
144: In this paper we deal with two external fields: 
145: the electric field of the photon (E1 transition amplitudes) 
146: and the weak field of the nucleus.     
147: In the RHF approximation the interaction between an external field 
148: $\hat{H}_{\rm ext}$ and atomic electrons is
149: $\langle \psi ^{HF}_{1}|\hat{H}_{\rm ext}|\psi ^{HF}_{2}\rangle$, 
150: where $\psi ^{HF}_{1}$ and $\psi ^{HF}_{1}$ are RHF orbitals.  
151: Inclusion of the polarization of the atomic core by an external field is  
152: reduced to the addition of a correction $\delta \hat{V}$ 
153: (which is the correction to the Hartree-Fock potential due to the interaction 
154: between the core and the external field) 
155: to the operator which describes the interaction, 
156: $\langle \psi ^{HF}_{1}|\hat{H}_{\rm ext}+
157: \delta \hat{V}|\psi ^{HF}_{2}\rangle$.
158: To include ``Brueckner-type'' correlation corrections the RHF orbitals are 
159: simply replaced by Brueckner ones, 
160: $\langle \psi ^{Br}_{1}|\hat{H}_{\rm ext}+
161: \delta \hat{V}|\psi ^{Br}_{2}\rangle$. 
162: The Brueckner-type correlations give the dominant corrections to the 
163: RHF approximation. They correspond to diagrams in which the 
164: interactions occur in the external lines of the self-energy operator 
165: (see, e.g., Fig. \ref{fig:pncdom}). 
166: Those diagrams in which the E1 interaction occurs in the internal 
167: lines are known as ``structural radiation'', 
168: while those in which the weak interaction occurs in the internal lines are 
169: known as the ``weak correlation potential''
170: (see, e.g., Fig. \ref{fig:pncint}).  
171: There is also a correction to the amplitudes arising from the 
172: normalization of states \cite{flam87}. 
173: The structural radiation, weak correlation potential, and 
174: normalization contributions are suppressed by the small parameter 
175: $E_{\rm ext}/E_{\rm core}\sim 1/10$, where  $E_{\rm ext}$ and 
176: $E_{\rm int}$ are excitation energies of the external and core electrons,
177: respectively.
178: 
179: The nuclear spin-independent weak interaction of an electron with 
180: the nucleus is
181: \begin{equation}
182: \hat{H}_{W}=-\frac{G_{F}}{2\sqrt{2}}\rho (r)Q_{W}\gamma _{5}
183: \end{equation}
184: where $G_{F}$ is the Fermi constant, $Q_{W}$ is the weak charge of the 
185: nucleus, $\gamma _{5}$ is a Dirac matrix, and $\rho (r)$ is the density 
186: of the nucleus.
187: Parity nonconserving E1 transition amplitudes, arising due to the 
188: simultaneous interaction of atomic electrons with the nuclear weak charge 
189: and the photon field, can be calculated using two methods:
190: from a mixed-states approach; 
191: or from a sum-over-states approach, 
192: in which experimental values (energies and E1 transition amplitudes) 
193: can be explicitly included.
194: Contributions to PNC E1 transition amplitudes are presented diagrammatically  
195: in Figs. \ref{fig:pncdom}, \ref{fig:pncint}.
196: 
197: In the mixed-states approach the PNC E1 transition amplitude between the 
198: states $ns$ and $(n-1)d_{3/2}$, $n=6$ for Cs and Ba$^{+}$, 
199: $n=7$ for Fr and Ra$^{+}$, is given by 
200: \begin{equation}
201: \label{eq:mixed}
202: E1_{PNC}=\langle \psi _{(n-1)d}|\hat{H}_{E1}+\delta \hat{V}_{E1}|
203: \delta \psi _{ns}\rangle +
204: \langle \psi _{(n-1)d}|\hat{H}_{W}+\delta \hat{V}_{W}|
205: \tilde{\psi} _{ns}\rangle +
206: \langle \psi _{(n-1)d}|\delta \hat{V}_{E1W}|\psi _{ns}\rangle ,
207: \end{equation}
208: where $\delta \psi $ and $\delta \hat{V}_{W}$ are corrections to 
209: single-electron wavefunctions and the Hartree-Fock potential caused 
210: by the weak interaction, $\tilde{\psi}$ and $\delta \hat{V}_{E1}$ are 
211: corrections to wavefunctions and the Hartree-Fock potential caused 
212: by the electric field of the photon, $\delta \hat{V}_{E1W}$ is the 
213: correction to the core potential due to the simultaneous action of the 
214: weak field and the electric field of the photon;  
215: the wavefunctions $\psi _{(n-1)d}$ and $\psi _{ns}$ correspond to Brueckner 
216: orbitals, and the corrections $\delta \psi$ and $\tilde{\psi}$ are found 
217: by solving the equations:
218: \begin{eqnarray}
219: (\hat{H}_{0}+\hat{\Sigma}-\epsilon)\delta \psi &=&
220: -(\hat{H}_{W}+\delta \hat{V}_{W})\psi \nonumber \\
221: (\hat{H}_{0}+\hat{\Sigma}-\epsilon)\tilde{\psi}&=&
222: -(\hat{H}_{E1}+\delta \hat{V}_{E1})\psi  \nonumber .
223: \end{eqnarray} 
224: This method is equivalent to calculating the diagrams presented 
225: in Fig. \ref{fig:pncdom} (with $\hat{\Sigma}$ chained to all orders, 
226: Fig. \ref{fig:sechain}) with the inclusion of the core polarization 
227: diagrams presented in Fig.~\ref{fig:corepol}.
228: 
229: Parity nonconserving E1 transition amplitudes between the states
230: $ns$ and $(n-1)d$ in the sum-over-states approach have the form 
231: \begin{eqnarray}
232: \label{eq:sos}
233: E1_{PNC}&=
234: &\sum _{n'}\frac{\langle (n-1)d_{3/2}|\hat{H}_{E1}+\delta \hat{V}_{E1}
235: |n'p_{1/2}\rangle 
236: \langle n'p_{1/2}|\hat{H}_{W}+\delta \hat{V}_{W}|ns\rangle}
237: {E_{ns}-E_{n'p_{1/2}}} \nonumber \\ 
238: &&+
239: \sum _{n'}\frac{\langle (n-1)d_{3/2}|\hat{H}_{W}+\delta \hat{V}_{W}
240: |n'p_{3/2}\rangle 
241: \langle n'p_{3/2}|\hat{H}_{E1}+\delta \hat{V}_{E1}|ns\rangle}
242: {E_{nd_{3/2}}-E_{n'p_{3/2}}},
243: \end{eqnarray}
244: where the sum is taken over a complete set of $p_{1/2}$ and $p_{3/2}$ states.
245: 
246: Note that the sum-over-states approach should also include the states with 
247: double-excitations like, for example in Cs, 
248: $\langle 5p^{6}6s|\hat{H}_{W}|5p^{5}6p7s\rangle$.
249: In the mixed-states approach these states are included, for example, 
250: in the last term of Eq. \ref{eq:mixed} 
251: (see also diagram (c) of Fig. \ref{fig:corepol}).
252: These exotic states contribute due to their mixing 
253: with the single-excited electron states.
254: This means that the mixed-states calculation (\ref{eq:mixed}) 
255: is more complete than the sum-over-states (\ref{eq:sos}) 
256: {\it unless} the high-energy states with two or more excited electrons 
257: are included into the sum. 
258: 
259: However, the accuracy of pure {\it ab initio} calculations for $s$-$d$ 
260: transitions is not very good because of the huge correlations for $d$-states. 
261: On the other hand, we will see in the next section that this problem can 
262: be avoided in the sum-over-states approach by using experimental values 
263: for the $p$-$d$ E1 transition amplitudes. 
264: Therefore, the best accuracy can be achieved when both methods are combined. 
265: Substitution of experimental values into the sum-over-states 
266: approach leads to a correction to the PNC amplitude which can be added to 
267: the mixed-states result. Following this procedure, it is possible to 
268: determine the PNC $s$-$d$ amplitudes with an accuracy of about $1\%$ 
269: (see discussion in the end of the next section). 
270: 
271: \section{Results}
272: 
273: Hartree-Fock energies for Cs, Fr, Ba$^{+}$, and Ra$^{+}$ are presented in 
274: Table~\ref{tab:energies}. These have an accuracy of the order of 10\%.   
275: The Brueckner energies, including the three series of higher order diagrams, 
276: are also presented in Table~\ref{tab:energies}.
277: These energies have an accuracy of the order of 0.1\%. 
278: 
279: Electric dipole transition amplitudes between the states $m$ and $m'$ are 
280: calculated in length form, 
281: $\langle m'||\hat{H}_{E1}||m\rangle = \langle m'||{\bf r}||m\rangle 
282: =C_{mm'}R$, where $C_{mm'}$ are angular coefficients and 
283: $R$ is the radial integral. 
284: In Table \ref{tab:E1} we present radial integrals relevant to the 
285: sum-over-states calculation for Cs, Fr, Ba$^{+}$, and Ra$^{+}$.
286: In this table we present the values obtained in the RHF approximation and 
287: show the contribution of core polarization 
288: to the RHF integrals; we also present the (unfitted) Brueckner 
289: results and the contributions arising from structural radiation and 
290: normalization of states.
291: In Table \ref{tab:moreE1} we present radial integrals 
292: between the lower states of the four atoms which are calculated with 
293: fitted Brueckner orbitals and with structural radiation and normalization 
294: contributions included. 
295: Experimental values for Cs and Fr are presented in Table \ref{tab:expe1}.
296: The Cs transitions $5d_{3/2}$-$6p_{3/2}$ and $5d_{3/2}$-$6p_{1/2}$
297: were extracted from the measurement \cite{cstau} of the $5d_{3/2}$ lifetime, 
298: $\tau =909(15)$~ns by assuming that the ratio of the calculated radial 
299: integrals corresponds to the ratio of the experimental values. 
300: This assumption was also used to obtain the Fr $7d_{3/2}$-$7p_{3/2}$ and 
301: $7d_{3/2}$-$7p_{1/2}$ radial integrals from the 
302: measured lifetime $\tau =73.6(3)$~ns \cite{frdtau}. 
303: With the exception of the Cs $6s$-$7p_{1/2}$ transition, 
304: the calculations of $s$-$p_{1/2}$ radial integrals agree with 
305: experiment at the level of 0.1\%. 
306: The poor accuracy of the $6s$-$7p_{1/2}$ radial integral is due to the 
307: fact that the main RHF contribution is very small and the relative 
308: contribution of all corrections is large. 
309: The $5d$-$6p$ radial integrals for Cs have poor accuracy, 
310: deviating from experiment by about 4\%. This is indicative 
311: of the poor calculation of $d$-states due to very large correlation 
312: corrections. The accuracy for the Fr $7d$-$7p$ radial integrals is 
313: about 1\%. The reason that this accuracy is better than that for
314: Cs $5d$-$6p$ is because the accuracy of the higher $d$-levels 
315: (here $7d$ rather than $6d$) is better due to smaller correlation 
316: corrections.
317: 
318: We have calculated the lifetimes of the low-lying states of 
319: Ba$^{+}$ and Ra$^{+}$.
320: The $nd_{3/2}$ states of Ba$^{+}$ ($n=6$) and Ra$^{+}$ ($n=7$) decay 
321: directly to the ground state via the E2 transition; 
322: the $nd_{5/2}$ states decay via both the E2 and M1 transitions.
323: Lifetimes of the Ba$^{+}$ $5d$ states are presented in 
324: Table~\ref{tab:lifetime}.
325: The calculations were performed with fitted Brueckner orbitals; 
326: core polarization, structural radiation, and normalization contributions 
327: were included into the E2 transition amplitudes. 
328: The calculations are in good agreement with experiment.
329: We have also presented the calculations performed by 
330: Guet and Johnson \cite{G&J}.
331: It appears that for the state $5d_{5/2}$ they have not taken 
332: into account the M1 transition. This seems to be the reason for the 
333: discrepancy between the lifetime calculations for this state. 
334: From our calculations it is seen that inclusion of the M1 transition 
335: effectively decreases the lifetime of the $5d_{5/2}$ state from $36.3$~s 
336: to $30.3$~s.      
337: For Ra$^{+}$ we obtained the lifetimes 
338: $\tau =0.641$~s and $\tau =0.302$~s for the states
339: $6d_{3/2}$ and $6d_{5/2}$, respectively; these were calculated in the same 
340: way as for the Ba$^{+}$ $5d$ states.
341: The lifetimes for states of Ra$^{+}$ have not been measured.
342: Lifetimes of all other states are strongly dominated by E1 transitions 
343: and so can be calculated using the radial integrals presented in 
344: Table~\ref{tab:moreE1}.
345: The calculated lifetimes of the $6p$ states of Ba$^{+}$ are in excellent 
346: agreement with experiment and calculations by Guet and Johnson 
347: (see Table~\ref{tab:lifetime}).
348: 
349: The mixed-states results for the E1 PNC transition amplitudes are listed 
350: in Table~\ref{tab:mixed}.
351: The results of the sum-over-states calculation, 
352: and the contributions of the six terms corresponding to the summation 
353: of the $n$-$(n+2)~p$ states, are presented in Table~\ref{tab:sos}.
354: In both calculations the contributions of structural radiation, 
355: weak correlation potential, and normalization of states are not included.
356: The $s$-$d$ PNC amplitudes are up to about four times as 
357: large as their corresponding $s$-$s$ amplitudes.
358: Furthermore, unlike the contributions to the sum-over-states calculation 
359: in Cs $6s$-$7s$, in which the dominant contribution is about twice as large 
360: as the final result due to strong cancelations between three major terms in 
361: the sum, the PNC $s$-$d$ transitions in Cs, Fr, Ba$^{+}$, and Ra$^{+}$ 
362: are strongly dominated by a single term.
363: In each case this term corresponds to
364: $\langle (n-1)d_{3/2}|\hat{H}_{E1}|np_{1/2}\rangle 
365: \langle np_{1/2}|\hat{H}_{W}|ns\rangle /(E_{ns}-E_{np_{1/2}})$;
366: this term is different from the sum by less than 20\%.
367: 
368: Because the Cs $6p_{1/2}$-$5d_{3/2}$ E1 transition and energies are known
369: we can correct the mixed-states PNC result. Replacing the calculated 
370: values by these experimental values in the dominating term of the 
371: sum (\ref{eq:sos}) for Cs increases this term (and the total sum) by about
372: $4\%$. This correction is mostly due to the difference between the calculated 
373: and experimental E1 amplitude (see Table \ref{tab:expe1}).
374: From the $1\%$ accuracy of calculations of hyperfine structure constants 
375: for $s$ and $p_{1/2}$ states \cite{flam89e1hfs} we can expect that the accuracy 
376: of the $s$-$p_{1/2}$ weak matrix elements in this calculation is also about $1\%$.
377: Therefore, we can say that the uncertainty in the calculated {\it ab initio} 
378: $s$-$d$ $E1_{PNC}$ amplitudes is dominated by the uncertainty of the $p$-$d$ 
379: E1 matrix elements and constitutes about $4\%$ for Cs and about the same 
380: value, or a little more, for other atoms.
381: 
382: The correction to the $6s$-$5d$  $E1_{PNC}$ amplitude in Cs discussed in the 
383: previous paragraph is $0.126~iea_{B}(-Q_{W}/N)$. When it is added to the 
384: mixed-states result, the new value is 
385: $E1_{PNC}(6s-5d)=3.75~iea_{B}(-Q_{W}/N)$.
386: Since using the experimental $6p$-$5d$ E1 amplitude removes the main source 
387: of uncertainty, the accuracy of the modified 
388: result must be considerably better than $4\%$. Assuming high accuracy of 
389: $p$-$d$ transition amplitudes, one can say that the uncertainty is now 
390: dominated by the uncertainty of calculated $s$-$p$ weak matrix elements 
391: which is about $1\%$ (however, we believe that this accuracy can be improved 
392: beyond $1\%$ with the inclusion of weak correlation potential and 
393: normalization contributions). 
394: More rigorous calculations and a more detailed analysis of the accuracy 
395: will be carried out when the need arises from the progress in experiments. 
396: We expect that calculations of $E1_{PNC}$ amplitudes for Cs, Fr, Ba$^{+}$, and 
397: Ra$^{+}$, with empirical corrections, can reach an accuracy of about $1\%$.
398: 
399: \section{Conclusion}
400: 
401: We have calculated the PNC E1 transition amplitudes 
402: between $s$-$d$ states of Cs, Fr, Ba$^{+}$, and Ra$^{+}$.
403: Generally, high accuracy cannot be reached in purely {\it ab initio} 
404: calculations of these transitions due to the poor accuracy of $d$-states.
405: However, we have shown from a sum-over-states calculation that, unlike 
406: the Cs $6s$-$7s$ transition, the $s$-$d$ transitions we have mentioned 
407: are strongly dominated by a single term in the sum.
408: Moreover, this term corresponds to an $s$-$p_{1/2}$ weak matrix element, 
409: which can be calculated with an accuracy of better than $1\%$, 
410: and a $p_{1/2}$-$d_{3/2}$ E1 transition amplitude, which can 
411: be taken from experiment.
412: The need to reach high accuracy for $d$ states is therefore avoided.
413: In addition to this, PNC $s$-$d$ transitions 
414: are larger than the corresponding $s$-$s$ transitions. 
415: The mixed-states calculation can be modified by correcting the 
416: terms in the sum-over-states by inserting experimental E1 transitions and 
417: energies. If $p_{1/2}$-$d_{3/2}$ E1 transition amplitudes are measured 
418: to high accuracy, we believe that the accuracy of the calculations of 
419: $s$-$d$ PNC transitions for Cs, Fr, Ba$^{+}$, and Ra$^{+}$ can reach 1\%.   
420: 
421: \acknowledgments
422: 
423: We are grateful to N. Fortson for useful discussions. 
424: This work was supported by the Australian Research Council. 
425: 
426: %****************************************************************
427: \begin{thebibliography}{20}
428: 
429: \bibitem{wood}
430: 
431: C.S. Wood, S.C. Bennett, D. Cho, B.P. Masterson, J.L. Roberto, 
432: C.E. Tanner, and C.E. Wieman, Science {\bf 275}, 1759 (1997).
433: 
434: \bibitem{flam89pnc}
435: 
436: V.A. Dzuba, V.V. Flambaum, and O.P. Sushkov, Phys. Lett. A {\bf 141},
437: 147 (1989).
438: 
439: \bibitem{johns90}
440: 
441: S.A. Blundell, W.R. Johnson, and J. Sapirstein, Phys. Rev. Lett.
442: 65, 1411 (1990); Phys. Rev. D {\bf 45}, 1602 (1992). 
443: 
444: \bibitem{bw}
445: 
446: S.C. Bennett and C.E. Wieman, Phys. Rev. Lett. {\bf 82},
447: 2484 (1999); {\bf 82}, 4153 (1999); {\bf 83}, 889 (1999). 
448: 
449: \bibitem{Der}
450: 
451: A. Derevianko, Phys. Rev. Lett. {\bf 85}, 1618 (2000).
452: 
453: \bibitem{DzJohns}
454: 
455: V.A. Dzuba, C. Harabati, W.R. Johnson, and M.S. Safronova;
456: submitted to Phys. Rev. A.
457: 
458: \bibitem{casal}
459: 
460: R. Casalbuoni, S. De Curtis, D. Dominici, and R. Gatto, 
461: Phys. Lett. B {\bf 460}, 135 (1999).
462: 
463: \bibitem{rosner}
464: 
465: J.L. Rosner, Phys. Rev. D {\bf 61}, 016006 (1999).
466: 
467: \bibitem{erler}
468: 
469: J. Erler and P. Langacker, Phys. Rev. Lett. {\bf 84}, 212 (2000).
470: 
471: \bibitem{fortson}
472: 
473: N. Fortson, Phys. Rev. Lett. {\bf 70}, 2383 (1993); 
474: private communication.
475: 
476: \bibitem{flam87}
477: 
478: V.A. Dzuba, V.V. Flambaum, P.G. Silvestrov, and O.P. Sushkov, 
479: J. Phys. B {\bf 20}, 1399 (1987).	
480: 
481: \bibitem{flam89energy}
482: 
483: V.A. Dzuba, V.V. Flambaum, and O.P. Sushkov, 
484: Phys. Lett. A {\bf 140}, 493 (1989).
485: 
486: \bibitem{flam89e1hfs}
487: 
488: V.A. Dzuba, V.V. Flambaum, A.Ya. Kraftmakher, and O.P. Sushkov, 
489: Phys. Lett. A {\bf 142}, 373 (1989).
490: 
491: 
492: %-----------------------
493: %experimental energies
494: 
495: \bibitem{moore}
496: 
497: C.E. Moore, Atomic Energy Levels, Natl. Bur. Stand. Ref. Data Ser., 
498: Natl. Bur. Stand. (U.S.) Circ. No. 35 
499: (U.S. GPO, Washington, D.C., 1971), Vol. III.
500: 
501: \bibitem{Fr1}
502: 
503: J. Bauche {\it et al}, J. Phys. B {\bf 19}, L593 (1986).
504: 
505: \bibitem{Fr2}
506: 
507: H.T. Duong {\it et al}, Europhys. Lett. {\bf 3}, 175 (1987).
508: 
509: \bibitem{Fr3}
510: 
511: S.V. Andreev, V.S. Letokhov, and V.I. Mishin, 
512: Phys. Rev. Lett. {\bf 59}, 1274 (1987); 
513: S.V. Andreev, V.I. Mishin, and V.S. Letokhov, 
514: J. Opt. Soc. Am. B {\bf 5}, 2190 (1988).
515: 
516: \bibitem{Fr4}
517: 
518: E. Arnold {\it et al}, J. Phys. B {\bf 23}, 3511 (1990).
519: 
520: \bibitem{Fr5}
521: 
522: J.E. Simsarian, W. Shi, L.A. Orozco, G.D. Sprouse, and W.Z. Zhao,
523: Opt. Lett. {\bf 21}, 1939 (1996).
524: 
525: \bibitem{Fr6}
526: 
527: J.E. Simsarian, W.Z. Zhao, L.A. Orozco, and G.D. Sprouse,
528: Phys. Rev. A {\bf 59}, 195 (1999).
529: 
530: \bibitem{Fr7}
531: 
532: J.M. Grossman {\it et al}, Phys. Rev. A {\bf 62}, 052507 (2000).
533: 
534: %-----------------------
535: %radial integrals
536: %Cs
537: 
538: \bibitem{tanner99}
539: 
540: R.J. Rafac, C.E. Tanner, A.E. Livingston, and H.G. Berry, 
541: Phys. Rev. A {\bf 60}, 3648 (1999).
542: 
543: \bibitem{shab79}
544: 
545: L.N. Shabanova, Yu. M. Monakov, and A.M. Khlustalov, 
546: Opt. Spektrosk. {\bf 47}, 3 (1979).
547: 
548: \bibitem{bouchiat84}
549: 
550: M.-A. Bouchiat, J. Guena, and L. Pottier, 
551: J. Phys. (Paris) Lett. {\bf 45}, L523 (1984).
552: 
553: \bibitem{brw99}
554: 
555: S. C. Bennett, J. L. Roberts, and C. E. Wieman,
556: Phys. Rev. A {\bf 59}, R16 (1999).
557: 
558: \bibitem{cstau}
559: 
560: D. DiBerardino, C.E. Tanner, and A. Sieradzan, 
561: Phys. Rev. A {\bf 57}, 4204 (1998).
562: 
563: %Fr
564: 
565: \bibitem{frptau}
566: 
567: J.E. Simsarian, L.A. Orozco, G.D. Sprouse, and W.Z. Zhao, 
568: Phys. Rev. A {\bf 57}, 2448 (1998).
569: 
570: \bibitem{frdtau}
571: 	
572: J.M. Grossman {\it et al}, Phys. Rev. A {\bf 62}, 062502 (2000).
573: 
574: %-----------------------
575: %lifetimes of Ba+ and Ra+ d-states
576: 
577: \bibitem{G&J}
578: 
579: C. Guet and W.R. Johnson, Phys. Rev. A {\bf 44}, 1531 (1991).
580: 
581: \bibitem{pinn95}
582: 
583: E.H. Pinnington, R.W. Berends, and M. Lumsden, 
584: J. Phys. B {\bf 28}, 2095 (1995).
585: 
586: \bibitem{N&D}
587: 
588: N. Yu, W. Nagourney, and H. Dehmelt, Phys. Rev. Lett. {\bf 78}, 4898 (1997).
589: 
590: \bibitem{MS90}
591: 
592: A.A. Madej and J.D. Sankey, Phys. Rev. A {\bf 41}, 2621 (1990).
593: 
594: 
595: \end{thebibliography}
596: 
597: %****************************************************************
598: \begin{table}
599: \caption{Energy levels (ionization potentials) of the lower states of 
600: Cs, Fr, Ba$^+$ and Ra$^+$ in units $-$cm$^{-1}$.}
601: \label{tab:energies}
602: \begin{tabular}{lcccccccc}
603:  && & Cs & && & Ba$^+$ &\\
604: State && RHF & Brueckner & Experiment \tablenotemark[1] & 
605:  & RHF & Brueckner & Experiment \tablenotemark[1]\\
606: $6s$ && 27954 & 31420 & 31407 && 75339 & 80813 & 80687 \\
607: $7s$ && 12112 & 12851 & 12871 && 36852 & 38333 & 38332 \\
608: $8s$ && 6793 & 7082 & 7090 && 22023 & 22651 & 22662 \\ 
609: $6p_{1/2}$ && 18790 & 20275 & 20228 && 57265 & 60581 & 60425 \\
610: $7p_{1/2}$ && 9223 & 9643 & 9641 && 30240 & 31332 & 31297 \\
611: $8p_{1/2}$ && 5513 & 5701 & 5698 && 18848 & 19378 & 19351 \\ 
612: $6p_{3/2}$ && 18389 & 19708 & 19674& & 55873 & 58860 & 58734 \\
613: $7p_{3/2}$ && 9079 & 9460 & 9460 && 29699 & 30704 & 30676 \\
614: $8p_{3/2}$ && 5446 & 5618 & 5615 && 18580 & 19075 & 19051 \\ 
615: $5d_{3/2}$ && 14138 & 17023 & 16907 && 68139 & 76402 & 75813 \\
616: $6d_{3/2}$ && 7920 & 8824 & 8818 && 33266 & 34740 & 34737 \\
617: $7d_{3/2}$ && 4965 & 5362 & 5359 && 20251 & 20871 & 20887 \\ 
618: $5d_{5/2}$ && 14163 & 16915 & 16810& & 67665 & 75525 & 75012 \\
619: $6d_{5/2}$ && 7921 & 8781 & 8775 && 33093 & 34536 & 34532 \\
620: $7d_{5/2}$ && 4963 & 5341 & 5338 && 20167 & 20777 & 20792 \\ 
621: \hline 
622:  && & Fr && & & Ra$^+$ & \\
623: State && RHF & Brueckner & Experiment \tablenotemark[2] & 
624:  & RHF & Brueckner & Experiment \tablenotemark[1]\\
625: $7s$ && 28768 & 32841 & 32849 && 75900 & 81960 & 81842 \\
626: $8s$ && 12282 & 13071 & 13116 && 36861 & 38405 & 38437 \\
627: $9s$ && 6858 & 7164 & 7178 && 22004 & 22659 & 22677 \\  
628: $7p_{1/2}$ && 18856 & 20674 & 20612 && 56879 & 60681 & 60491 \\
629: $8p_{1/2}$ && 9240 & 9730 & 9736 && 30053 & 31244 & 31236 \\
630: $9p_{1/2}$ && 5521 & 5737 & - && 18748 & 19332 & - \\
631: $7p_{3/2}$ && 17656 & 18944 & 18925 && 52906 & 55734 & 55633 \\
632: $8p_{3/2}$ && 8811 & 9180 & 9191 && 28502 & 29447 & 29450 \\
633: $9p_{3/2}$ && 5319 & 5486 & - && 17975  & 18462 & 18432 \\ 
634: $6d_{3/2}$ && 13826 & 16610 & - && 62356 & 70149 & 69758 \\
635: $7d_{3/2}$ && 7725 & 8583 & 8604 && 31575 & 33060 & 33098 \\
636: $8d_{3/2}$ && 4857 & 5241 & 5248 && 19451 & 20079 & 20107 \\ 
637: $6d_{5/2}$ && 13925 & 16413 & - && 61592 & 68449 & 68099 \\
638: $7d_{5/2}$ && 7747 & 8496 & 8516 && 31204 & 32569 & 32602 \\
639: $8d_{5/2}$ && 4863 & 5197 & 5203 && 19261 & 19849 & 19868 \\  
640: %
641: \end{tabular}
642: \tablenotetext[1]{Taken from \cite{moore}.}
643: \tablenotetext[2]{Measured in Refs. 
644: \cite{Fr1,Fr2,Fr3,Fr4,Fr5,Fr6,Fr7}.}
645: \end{table}
646: %****************************************************************
647: \begin{table}
648: \caption{Calculated radial integrals (a.u.) for Cs, Fr, Ba$^+$, Ra$^+$. 
649: We present RHF values, RHF with core polarization, the Brueckner 
650: result with core polarization included, and structural radiation 
651: and normalization of states; 0.0 signifies that the value is smaller 
652: than the number of figures specified.}
653: \label{tab:E1}
654: \begin{tabular}{ccccccc}
655: Atom &
656: Transition &
657: RHF &
658: RHF + &
659: Brueckner + &
660: Structural & 
661: Normal- \\
662:  & & & core polar- & core polar- & radiation & ization \\
663:  & & & ization & ization & & of states\\
664: \hline
665: Cs & $6s_{1/2} - 6p_{3/2}$ & -6.432 & -6.074 & -5.500 & 
666: -0.028 & 0.047 \\
667: &$6s_{1/2} - 7p_{3/2}$ & -0.602 & -0.440 & -0.463 & 
668: -0.013 & 0.003 \\
669: &$6s_{1/2} - 8p_{3/2}$ & -0.245 & -0.143 & -0.162 & 
670: -0.008 & 0.001 \\
671: &$6p_{1/2} - 5d_{3/2}$ & 7.775 & 7.481 & 6.050 & 
672: 0.026 & -0.050 \\
673: &$7p_{1/2} - 5d_{3/2}$ & -3.498 & -3.591 & -1.742 &
674: 0.011 & 0.011 \\
675: &$8p_{1/2} - 5d_{3/2}$ & -0.860 & -0.916 & -0.556 &
676: 0.007 & 0.003 \\
677: &&&&&&\\
678: Fr & $7s_{1/2} - 7p_{3/2}$ & -6.140 & -5.739 & -5.128 &
679: -0.032 & 0.051 \\  
680: &$7s_{1/2} - 8p_{3/2}$ & -0.949 & -0.760 & -0.748 &
681: -0.015 & 0.006 \\ 
682: &$7s_{1/2} - 9p_{3/2}$ & -0.452 & -0.332 & -0.336 &
683: -0.009 & 0.003 \\
684: &$7p_{1/2} - 6d_{3/2}$ & 7.986 & 7.613 & 6.256 & 
685: 0.035 & -0.061 \\
686: &$8p_{1/2} - 6d_{3/2}$ & -4.005 & -4.116 & -2.249 &
687: 0.014 & 0.015 \\
688: &$9p_{1/2} - 6d_{3/2}$ & -0.941 & -1.008 & -0.704 &
689: 0.008 & 0.004 \\
690: &&&&&&\\
691: Ba$^+$ & $6s_{1/2} - 6p_{3/2}$ & -4.744 & -4.314 & -4.056 &
692: -0.032 & 0.046 \\ 
693: &$6s_{1/2} - 7p_{3/2}$ & -0.226 & -0.036 & -0.015 &
694: -0.013 & 0.0 \\
695: &$6s_{1/2} - 8p_{3/2}$ & -0.068 & 0.054 & 0.076 &
696: -0.007 & -0.001 \\
697: &$6p_{1/2} - 5d_{3/2}$ & 3.244 & 2.964 & 2.634 & 
698: 0.026 & -0.034 \\
699: &$7p_{1/2} - 5d_{3/2}$ & 0.304 & 0.184 & 0.225 &
700: 0.010 & -0.002 \\
701: &$8p_{1/2} - 5d_{3/2}$ & 0.169 & 0.092 & 0.099 &
702: 0.007 & -0.001 \\
703: &&&&&&\\
704: Ra$^+$ & $7s_{1/2} - 7p_{3/2}$ & -4.624 & -4.154 & -3.885 &
705: -0.035 & 0.048 \\ 
706: &$7s_{1/2} - 8p_{3/2}$ & -0.541 & -0.320 & -0.286 &
707: -0.015 & 0.003 \\
708: &$7s_{1/2} - 9p_{3/2}$ & -0.243 & -0.098 & -0.067 &
709: -0.009 & 0.001 \\
710: &$7p_{1/2} - 6d_{3/2}$ & 3.851 & 3.448 & 3.067 &
711: 0.037 & -0.047 \\
712: &$8p_{1/2} - 6d_{3/2}$ & 0.091 & -0.075 & 0.009 &
713: 0.014 & 0.0 \\
714: &$9p_{1/2} - 6d_{3/2}$ & 0.075 & -0.030 & -0.014 &
715: 0.008 & 0.0 \\
716: 
717: \end{tabular}
718: \end{table}
719: %****************************************************************
720: \begin{table}
721: \caption{Radial integrals (a.u.) for states of Cs, Fr, Ba$^{+}$ and Ra$^{+}$. 
722: Fitted Brueckner orbitals are used; core polarization, and 
723: structural radiation and normalization of states are also 
724: included.}
725: \label{tab:moreE1}
726: \begin{tabular}{cccccccc}
727: & & $6p_{1/2}$ & $7p_{1/2}$ & $8p_{1/2}$ & $6p_{3/2}$ & 
728: $7p_{3/2}$ & $8p_{3/2}$ \\
729: \hline
730: Cs & $6s$ & -5.508 & -0.313 & -0.081 & -5.482 & -0.471 & -0.171 \\ 
731: & $7s$ & 5.211 & -12.605 & -1.137 & 5.625 & -12.383 & -1.419 \\
732: & $8s$ & 1.266 & 11.386 & -21.753 & 1.273 & 12.163 & -21.252 \\
733: & $5d_{3/2}$ & 6.072 & -1.785 & -0.560 & 
734: 6.120 & -1.579 & -0.506 \\
735: & $6d_{3/2}$ & -3.696 & 15.558 & -4.234 & -4.082 & 15.612 & -3.726 \\
736: & $7d_{3/2}$ & -1.806 & -5.632 & 27.670 & -1.918 & -6.379 & 27.773 \\
737: & $5d_{5/2}$ & - & - & - & 6.190 & -1.641 & -0.522 \\
738: & $6d_{5/2}$ & - & - & - & -3.986 & 15.699 & -3.886 \\
739: & $7d_{5/2}$ & - & - & - & -1.893 & -6.186 & 27.876 \\
740: \hline
741: Ba$^{+}$ & $6s$ & -4.054 & 0.121 & 0.141 & -4.048 & -0.030 
742: & 0.063 \\ 
743: & $7s$ & 3.053 & -8.583 & -0.139 & 3.362 & -8.464 & -0.401 \\
744: & $8s$ & 0.863 & 6.080 & -14.153 & 0.888 & 6.634 & -13.884 \\
745: & $5d_{3/2}$ & 2.646 & 0.226 & 0.103 & 2.584 & 0.285 & 0.135 \\
746: & $6d_{3/2}$ & -4.234 & 7.488 & 0.101 & -4.520 & 7.311 & 0.282 \\
747: & $7d_{3/2}$ & -1.189 & -7.134 & 13.494 & -1.170 & -7.674 & 13.195 \\
748: & $5d_{5/2}$ & - & - & - & 2.658 & 0.279 & 0.133 \\
749: & $6d_{5/2}$ & - & - & - & -4.469 & 7.418 & 0.235 \\
750: & $7d_{5/2}$ & - & - & - & -1.1865 & -7.5570 & 13.352 \\
751: \hline
752: \hline
753: & & $7p_{1/2}$ & $8p_{1/2}$ & $9p_{1/2}$ & $7p_{3/2}$ & 
754: $8p_{3/2}$ & $9p_{3/2}$ \\
755: \hline
756: Fr & $7s$ & -5.242 & -0.331 & -0.093 & -5.107 & -0.748 & -0.343 \\ 
757: & $8s$ & 5.217 & -12.326 & -1.206 & 6.484 & -11.536 & -1.947 \\
758: & $9s$ & 1.256 & 11.428 & -21.382 & 1.215 & 13.777 & -19.660 \\
759: & $6d_{3/2}$ & 6.237 & -2.229 & -0.691 & 6.417 & -1.553 & -0.516 \\
760: & $7d_{3/2}$ & -3.014 & 15.893 & -5.488 & -4.186 & 16.175 & -3.829 \\
761: & $8d_{3/2}$ & -1.606 & -4.296 & 27.952 & -1.975 & -6.546 & 28.464 \\
762: & $6d_{5/2}$ & - & - & - & 6.576 & -1.684 & -0.549 \\
763: & $7d_{5/2}$ & - & - & - & -3.974 & 16.368 & -4.184 \\
764: & $8d_{5/2}$ & - & - & - & -1.917 & -6.106 & 28.695 \\
765: \hline
766: Ra$^{+}$ & $7s$ & -3.948 & 0.108 & 0.142 & -3.877 & -0.294 
767: & -0.082 \\
768: & $8s$ & 3.104 & -8.523 & -0.168 & 4.038 & -8.071 & -0.891 \\ 
769: & $9s$ & 0.867 & 6.167 & -14.122 & 0.897 & 7.825 & -13.137 \\
770: & $6d_{3/2}$ & 3.074 & 0.011 & -0.011 & 2.913 & 0.245 & 0.110 \\
771: & $7d_{3/2}$ & -3.774 & 8.262 & -0.412 & -4.662 & 7.801 & 0.256 \\
772: & $8d_{3/2}$ & -1.240 & -6.152 & 14.649 & -1.242 & -7.812 & 13.854 \\
773: & $6d_{5/2}$ & - & - & - & 3.109 & 0.224 & 0.105 \\
774: & $7d_{5/2}$ & - & - & - & -4.515 & 8.087 & 0.117 \\
775: & $8d_{5/2}$ & - & - & - & -1.281 & -7.476 & 14.276 \\
776: \end{tabular}
777: \end{table}
778: %****************************************************************
779: \begin{table}
780: \caption{Calculated (Table \ref{tab:moreE1}) and experimental 
781: radial integrals (a.u.).}
782: \label{tab:expe1}
783: \begin{tabular}{cccc}
784: Atom & Transition & Calc. & Exp.\\
785: \hline
786: Cs & $6s$-$6p_{1/2}$ & -5.508 & -5.497(8) \tablenotemark[1] \\
787:  & $6s$-$6p_{3/2}$ & -5.482 & -5.476(6) \tablenotemark[1] \\
788:  & $6s$-$7p_{1/2}$ & -0.313 & -0.348(3) \tablenotemark[2] \\
789:  & $7s$-$6p_{1/2}$ & 5.211 & 5.185(27) \tablenotemark[3] \\
790:  & $7s$-$6p_{3/2}$ & 5.625 & 5.611(27) \tablenotemark[3] \\
791:  & $7s$-$7p_{1/2}$ & -12.605 & -12.625(18) \tablenotemark[4]\\
792:  & $7s$-$7p_{3/2}$ & -12.383 & -12.401(17) \tablenotemark[4]\\
793:  & $5d_{3/2}$-$6p_{1/2}$ & 6.072 & 6.31(5)\tablenotemark[5] \\
794:  & $5d_{3/2}$-$6p_{3/2}$ & 6.120 & 6.36(5)\tablenotemark[5] \\ 
795:  & $5d_{5/2}$-$6p_{3/2}$ & 6.190 & 6.40(2)\tablenotemark[5] \\
796: \hline
797: Fr & $7s$-$7p_{1/2}$ & -5.242 & -5.238(10) \tablenotemark[6]\\ 
798:  & $7s$-$7p_{3/2}$ & -5.107 & -5.108(13) \tablenotemark[6]\\
799:  & $7d_{3/2}$-$7p_{1/2}$ & -3.014 & -3.05(1) \tablenotemark[7] \\
800:  & $7d_{3/2}$-$7p_{3/2}$ & -4.186 & -4.24(2) \tablenotemark[7] \\
801:  & $7d_{5/2}$-$7p_{3/2}$ & -3.974 & -4.02(8) \tablenotemark[7] \\
802: \end{tabular}
803: \tablenotetext[1]{Reference \cite{tanner99}.}
804: \tablenotetext[2]{Reference \cite{shab79}.}
805: \tablenotetext[3]{Reference \cite{bouchiat84}.}
806: \tablenotetext[4]{Reference \cite{brw99}.}
807: \tablenotetext[5]{Reference \cite{cstau}.}
808: \tablenotetext[6]{Reference \cite{frptau}.}
809: \tablenotetext[7]{Reference \cite{frdtau}.}
810: \end{table}
811: %****************************************************************
812: \begin{table}
813: \caption{Lifetimes of low-lying states of Ba$^+$.}
814: \label{tab:lifetime} 
815: \begin{tabular}{ccccc}
816: Atom & State & $\tau ^{\rm this ~ work}$ & 
817: $\tau $\tablenotemark[1] & $\tau ^{\rm expt}$ \\
818: \hline
819: Ba$^{+}$ & $6p_{1/2}$ & $7.89$~ns & $7.99$~ns & $7.90(10)$~ns 
820: \tablenotemark[2] \\
821: 	 & $6p_{3/2}$ & $6.30$~ns & $6.39$~ns & $6.32(10)$~ns 
822: \tablenotemark[2] \\
823: 	 & $5d_{3/2}$ & $81.5$~s & $83.7$~s & $79.8(4.6)$~s 
824: \tablenotemark[3]\\
825:          & $5d_{5/2}$ & $30.3$~s & $37.2$~s & $34.5(3.5)$~s \tablenotemark[4]
826: \end{tabular}
827: \tablenotetext[1]{Reference \cite{G&J}.} 
828: \tablenotetext[2]{Reference \cite{pinn95}.}
829: \tablenotetext[3]{Reference \cite{N&D}.}
830: \tablenotetext[4]{Reference \cite{MS90}.}
831: \end{table}
832: %****************************************************************
833: 
834: \begin{table}
835: \caption{Mixed-states results for PNC E1 transition amplitudes 
836: between the states $ns$-$(n-1)d$, $\langle (n-1)d|E1_{z}|ns\rangle$, 
837: where $n=6$ for Cs and Ba$^{+}$ and $n=7$ for Fr and Ra$^{+}$; units are
838: $10^{-11}iea_{B}(-Q_{W}/N)$.}
839: \label{tab:mixed}
840: \begin{tabular}{ccccc}
841:  & Cs & Fr & Ba$^{+}$ & Ra$^{+}$ \\
842: \hline
843: Mixed-states $E1_{PNC}$ & 3.62 & 57.1 & 2.17 & 42.9 \\
844: \end{tabular}
845: \end{table}
846: 
847: %****************************************************************
848: 
849: \begin{table}
850: \caption{Results of the sum-over-states calculations for the 
851: PNC E1 transition amplitudes for Cs, Fr, Ba$^{+}$, and Ra$^{+}$.
852: We present the contributions of the terms in the sum corresponding 
853: to the intermediate $n$-$(n+2)$$p$ states, the contribution due to 
854: all other intermediate $p$-states, and the total value; 
855: units $10^{-11}iea_{B}(-Q_{W}/N)$; 0.0 means that the term is 
856: smaller than the number of figures specified.}  
857: \label{tab:sos}
858: \begin{tabular}{cccc}
859: & & $\frac{\langle d|\hat{H}_{E1}|np_{1/2}\rangle 
860: \langle np_{1/2}|\hat{H}_{W}|s\rangle}{E_{s}-E_{np_{1/2}}}$ &
861: $\frac{\langle d|\hat{H}_{W}|np_{3/2}\rangle 
862: \langle np_{3/2}|\hat{H}_{E1}|s\rangle}{E_{d}-E_{np_{3/2}}}$ \\
863: \hline
864: Cs & n=6 & 3.154 & 0.728 \\
865:    & n=7 & -0.258 & -0.013 \\
866:    & n=8 & -0.047 & -0.002 \\
867:    & Other & \multicolumn{2}{c}{0.197}\\
868: \hline
869:    & Total & \multicolumn{2}{c}{3.76}\\
870: \hline
871: \hline
872: Fr & n=7 & 59.78 & 5.19 \\
873:    & n=8 & -6.13 & -0.15 \\
874:    & n=9 & -1.10 & -0.03 \\
875:    & Other & \multicolumn{2}{c}{1.95}\\
876: \hline
877:    & Total & \multicolumn{2}{c}{59.5}\\
878: \hline
879: \hline
880: Ba$^{+}$ & n=6 & 2.036 & -0.264 \\
881:    & n=7 & 0.045 & -0.001 \\
882:    & n=8 & 0.012 & 0.0 \\
883:    & Other & \multicolumn{2}{c}{0.511}\\
884: \hline
885:    & Total & \multicolumn{2}{c}{2.34}\\
886: \hline
887: \hline
888: Ra$^{+}$ & n=7 & 40.69 & -2.33 \\
889:    & n=8 & 0.11 & -0.05 \\
890:    & n=9 & 0.02 & -0.01 \\
891:    & Other & \multicolumn{2}{c}{7.47}\\
892: \hline
893:    & Total & \multicolumn{2}{c}{45.9}\\
894: \end{tabular}
895: \end{table}
896: %****************************************************************
897: 
898: \center
899: \widetext
900: \input psfig
901: \psfull
902: 
903: \begin{figure}[b]
904: \centerline{\psfig{file=sdfeyncorr.eps,clip=}}
905: \caption{Second-order correlation diagrams for the valence 
906: electron ($\hat \Sigma$ operator). Dashed line is the Coulomb interaction. 
907: Loop is the polarization of the atomic core.}
908: \label{fig:2ndorder}
909: \end{figure}
910: 
911: \begin{figure}[b]
912: \centerline{\psfig{file=screening.eps,clip=}}
913: \caption{Screening of the Coulomb interaction.}
914: \label{fig:screening}
915: \end{figure}
916: 
917: \begin{figure}[b]
918: \centerline{\psfig{file=hpchain.eps, clip=}}
919: \caption{Hole-particle interaction in the polarization operator.}
920: \label{fig:hpchain}
921: \end{figure}
922: 
923: \begin{figure}[b]
924: \centerline{\psfig{file=hpscreense.eps, clip=}}
925: \caption{The electron self-energy operator with screening and hole-particle 
926: interaction included.}
927: \label{fig:hpscreense}
928: \end{figure}
929: 
930: \begin{figure}[b]
931: \centerline{\psfig{file=sechain.eps, clip=}}
932: \caption{Chaining of the self-energy operator.}
933: \label{fig:sechain}
934: \end{figure}
935: 
936: \begin{figure}[b]
937: \centerline{\psfig{file=pncdom.eps, clip=}}
938: \caption{Brueckner-type correlation corrections to the PNC E1 transition 
939: amplitude; 
940: the crosses denote the weak interaction and the 
941: dashed lines denote the electromagnetic interaction.}
942: \label{fig:pncdom}
943: \end{figure}
944: 
945: \begin{figure}[b]
946: \centerline{\psfig{file=pncint.eps, clip=}}
947: \caption{Small corrections to the PNC E1 transition amplitude: 
948: external field inside the correlation potential.
949: In diagrams (a) the weak interaction is inside the 
950: correlation potential ($\delta \Sigma$ denotes the change in $\Sigma $ due to 
951: the weak interaction); this is known as the weak correlation potential.
952: Diagrams (b,c) represent structural radiation 
953: (photon field inside the correlation potential).
954: In diagram (b) the weak interaction occurs in the external lines;
955: in diagram (c) both the weak and electromagnetic interactions 
956: occur in the internal lines.}
957: \label{fig:pncint}
958: \end{figure}
959: 
960: \begin{figure}[b]
961: \centerline{\psfig{file=corepol.eps, clip=}}
962: \caption{Examples of diagrams representing the polarization of the 
963: atomic core by external fields. 
964: (The diagrams we have presented are exchange diagrams; 
965: there are also direct diagrams.)
966: In diagrams (a) and (b) the core is 
967: polarized by a single field (the dashed line denotes the E1 
968: interaction and the cross denotes the weak interaction). 
969: Diagram (c) corresponds to the polarization of the core by both fields. }
970: \label{fig:corepol}
971: \end{figure}
972: 
973: \end{document}
974: