nucl-th0509008/nib.tex
1: \tolerance = 1000
2: \documentclass[preprint,aps,showpacs]{revtex4}
3: \usepackage{graphicx}
4: \begin{document}
5: 
6: \title{Sensitivity to multi-phonon excitations in heavy-ion fusion reactions}
7: \author{H. Esbensen}
8: \affiliation{Physics Division, Argonne National Laboratory, Argonne, Illinois 60439}
9: \date{\today}
10: 
11: \begin{abstract}
12: Measured cross sections for the fusion of $^{64}$Ni with $^{64}$Ni, $^{74}$Ge, 
13: and $^{100}$Mo targets are analyzed in a coupled-channels approach. 
14: The data for the $^{64}$Ni target above 0.1 mb are reproduced by including 
15: couplings to the low-lying $2^+$ and $3^-$ states and the mutual 
16: and two-phonon excitations of these states.
17: The calculations become more challenging as the fusing nuclei become 
18: softer and heavier, and excitations to multi-phonon states start to 
19: play an increasingly important role.
20: Thus it is necessary to include up to four-phonon excitations in order to 
21: reproduce the data for the $^{64}$Ni+$^{74}$Ge system. Similar calculations 
22: for $^{64}$Ni+$^{100}$Mo, and also for the symmetric $^{74}$Ge+$^{74}$Ge system, 
23: show large discrepancies with the data.
24: Possible ways to improve the calculations are discussed.
25: \end{abstract}
26: \pacs{24.10.Eq, 25.70.Jj}
27: \maketitle
28: 
29: \section{Introduction}
30: 
31: Heavy-ion fusion data have in many cases been reproduced fairly well at 
32: energies close to the Coulomb barrier by coupled-channels calculations.
33: The best agreement is achieved for lighter or asymmetric systems, whereas 
34: the fusion of very heavy systems poses a serious challenge \cite{baha}.
35: Recently it was realized that the data for heavy systems 
36: are suppressed compared to single-channel calculations at energies 
37: above the Coulomb barrier \cite{newt1} and the suppression even 
38: persists in comparison to coupled-channels calculations \cite{newt2}.
39: There are also challenges in the theoretical description of heavy-ion 
40: fusion reactions at energies far below the Coulomb barrier, where
41: the measured cross sections are hindered compared to coupled-channels 
42: calculations \cite{syst}. While the suppression at high energy is 
43: largest for heavy systems, the hindrance of fusion far below the 
44: barrier appears to be a more general phenomenon. 
45: 
46: The structure input to coupled-channels calculations is commonly based 
47: on a vibrational or rotational model, with parameters determined from 
48: the lowest $2^+$ or $3^-$ excitations. These models make it possible
49: to calculate the coupling matrix elements to all orders in the
50: deformation amplitudes \cite{eslan,Takiful}, but they may not always provide 
51: a realistic representation of the actual structure of the reacting nuclei.
52: It is therefore desirable to be able to extract the necessary structure 
53: input from other measurements.  The most detailed information is available 
54: for quadrupole excitations up to very high spins, but there are too many 
55: states to consider in a practical coupled-channels calculation.  
56: One way to simplify the calculations is to construct effective multi-phonon 
57: states. This has been done for two-phonon quadrupole excitations \cite{alge}, 
58: and the procedure will be expanded here to the three-phonon level.
59: 
60: In view of the problems mentioned above it is of interest to make a 
61: systematic coupled-channels analysis of heavy-ion fusion reactions, 
62: ranging from lighter to the heavier systems, so that one can see exactly how 
63: the calculations fail to reproduce the data as the system gets heavier.
64: It is also important to know which and how many states are actually needed 
65: to make the calculations converge. For example, it is well known that  
66: couplings to high-lying states, like the giant resonances, produce an overall 
67: energy shift of the calculated fusion cross section, essentially without 
68: affecting the shape of the energy dependent cross section \cite{Takipol}.
69: If that is the case, then there is no need to include such high-lying states 
70: because their effect can be compensated by adjusting the radius or the depth 
71: of the ion-ion potential.
72: 
73: The basic coupled-channels calculations that will be presented  
74: include couplings to the one- and two-phonon excitations and to the mutual 
75: excitations of the lowest $2^+$ and $3^-$ states in projectile and target.
76: Such calculations reproduce the fusion cross sections rather well for lighter 
77: systems \cite{alge,eslan}.  The calculations will be expanded 
78: to include up to three- or four-phonon excitations, and they will be used
79: to analyze the data for the fusion of $^{64}$Ni with 
80: $^{64}$Ni \cite{nini}, $^{74}$Ge \cite{Beck}, and $^{100}$Mo \cite{nimo},
81: and also the fusion data for $^{74}$Ge+$^{74}$Ge \cite{Begege}.  
82: The projectiles and targets are fairly neutron rich in these systems so the 
83: couplings to transfer channels should have a modest effect on fusion.  
84: 
85: \section{Details of the calculations} 
86: 
87: The coupled-channels calculations are performed as described in Ref. 
88: \cite{alge}.  The main assumption is the rotating frame, - or isocentrifugal 
89: approximation. A basic feature of this approximation is that the magnetic 
90: quantum number $M$ is a conserved quantity. Thus if we start with two 
91: spin-zero nuclei in the entrance channel, then only the $M=0$ component 
92: of the excited states will be populated. That leads to a large reduction
93: in the number of channels one has to include, in comparison to a more
94: complete calculation.
95: 
96: The basic nuclear field that is required in the calculations described in
97: Ref. \cite{alge} is the ion-ion potential $U(r)$. The couplings to 
98: inelastic channels can be expressed in terms of radial derivatives of the 
99: ion-ion potential, and couplings up to second order in the deformation 
100: amplitudes $\alpha_{n\lambda\mu}$ will be included in the calculations. 
101: In the rotating frame approximation we only consider the $\mu=0$ component,
102: $\alpha_{n\lambda 0}$, so the nuclear interaction has form \cite{alge},
103: \begin{equation}
104: V(r,\alpha_{n\lambda 0}) = U(r) - 
105: \frac{dU(r)}{dr} \sum_{n\lambda} s_{n\lambda} + 
106: \frac{1}{2} \frac{d^2U(r)}{dr^2}
107: \Bigl[\bigl(\sum_{n\lambda} s_{n\lambda}\bigr)^2 -
108: \langle gs|\bigl(\sum_{n\lambda} s_{n\lambda}\bigr)^2 |gs\rangle\Bigr],
109: \end{equation} 
110: where $s_{n\lambda}=R_n\alpha_{n\lambda 0}\sqrt{(2\lambda+1)/(4\pi)}$ 
111: and $R_n$ is the nuclear radius.
112: The ion-ion potential $U(r)$ that will be used in the following is the 
113: empirical proximity-type potential developed in Ref. \cite{BW},
114: \begin{equation}
115: \label{BWP}
116: U(r) = - \frac{16\pi\gamma a R_1R_2}{R_1+R_2} \
117: \Bigl( 1 + \exp\Bigl[\frac{r-R_1-R_2-\Delta R}{a}\Bigr]\Bigr)^{-1}.
118: \end{equation}
119: Here the diffuseness $a$ depends on the mass numbers $A_n$ of the colliding 
120: nuclei, $1/a = 1.17[1+0.53(A_1^{-1/3}+A_2^{-1/3})]$ fm$^{-1}$,
121: and the nuclear radii are $R_n=1.2 A_n^{1/3}-0.09$ fm.
122: The parameter $\Delta R$ is zero in Ref. \cite{BW} but is introduced 
123: here as an adjustable parameter. Finally, the surface tension $\gamma$ 
124: is set to the constant value $\gamma$ = 0.95 MeV~fm$^{-2}$, whereas 
125: Ref. \cite{BW} included some isospin dependence in $\gamma$.
126: 
127: The fusion is simulated by ingoing wave boundary conditions at the radial
128: separation $r_{min}$, where the total potential in the elastic channel 
129: develops a minimum inside the Coulomb barrier. The nuclear potential in 
130: Eq. (1) is supplemented with a weak, short ranged, imaginary part $iW(r)$, 
131: where 
132: \begin{equation}
133: W(r) = \frac{-10 \ {\rm MeV}}{1+\exp([r-r_{min}]/a_W)},
134: \end{equation}
135: and $a_W$ is set to 0.2 fm. The effect of the imaginary potential
136: is to reduce certain oscillations in the calculated cross sections,
137: as discussed later on.
138: 
139: The one-phonon states that will be used are shown in Table I.
140: The coupling parameters are expressed in terms of the $\beta$-values, 
141: $\beta_\lambda^C$ and $\beta_\lambda^N$, for the Coulomb and nuclear 
142: fields, respectively. The Coulomb $\beta$-values are taken from the 
143: literature \cite{NNDC}.  The nuclear $\beta$-values are uncertain 
144: and they are often set equal to the Coulomb values. In this work a 
145: 10\% higher nuclear $\beta$-value will be used. This is justified 
146: in the case of $^{64}$Ni from the analysis of inelastic scattering 
147: data \cite{Flem}, but such information is not always available.
148: 
149: In the rotating frame we only consider matrix elements of the
150: $\mu=0$ component of the deformation amplitudes $\alpha_{\lambda\mu}$
151: between $M=0$ magnetic sub-states \cite{alge}. The matrix 
152: elements are therefore of the form
153: \begin{equation}
154: \langle nI0| \alpha_{\lambda 0} | n'I'0 \rangle =
155: \langle I'0 \ \lambda 0|I0\rangle \
156: \frac{ \langle nI|| \alpha_{\lambda} || n'I'\rangle}{\sqrt{2I+1}},
157: \end{equation}
158: where $|nI0\rangle$ denotes a state at the $n$-phonon level with spin $I$.
159: The reduced matrix element determines the B-value,
160: \begin{equation}
161: B(E\lambda,nI\rightarrow n'I') = 
162: \Bigl(\frac{3ZeR_C^\lambda}{4\pi}\Bigr)^2 \
163: \frac{|\langle nI|| \alpha_{\lambda} || n'I'\rangle|^2}{2I+1},
164: \end{equation}
165: where $R_C=1.2A^{1/3}$ is the Coulomb radius of the nucleus being excited.
166: General matrix elements of the type given in Eq. (4) can be used to construct 
167: the couplings to effective two- and three-phonon states as discussed below 
168: for quadrupole excitations.
169: 
170: The coupling between the one- and two-phonon quadrupole states,
171: i.~e., between the $2^+_1$ states and the $0_2^+$, $2_2^+$ and $4_1^+$ 
172: states, can in some cases be obtained from the literature \cite{NNDC}.
173: Moreover, the two-phonon states are sometimes close in energy 
174: so it is convenient to combine them into one effective two-phonon 
175: state and thereby reduce the number of channels.
176: The procedure is described in Ref. \cite{alge}. 
177: The result (in the rotating frame approximation) is that the square of 
178: the coupling between the one-phonon and the effective two-phonon state is 
179: \begin{equation}
180: \label{ph2cou} 
181: \langle {\rm 2ph}| \alpha_{20} | {\rm 1ph}\rangle^2 =
182: \Bigl(\frac{4\pi}{3ZeR_C^2}\Bigr)^2
183: \sum_{I=0,2,4} \langle 2020|I0\rangle^2 \ B(E2, I\rightarrow 2),
184: \end{equation}
185: where the sum is over the spin $I$ of the three two-phonon states,
186: and $B(E2,I\rightarrow 2)$ are the reduced transition probabilities 
187: for the decay to the $2^+_1$ state.  The average two-phonon excitation 
188: energy is estimated by the energy weighted sum
189: \begin{equation}
190: E_{2{\rm ph}}= \sum_{I=0,2,4} E_2(I) \ 
191: \frac{|\langle {\rm 2ph},I0| \alpha_{20}| {\rm 1ph}, 20\rangle|^2}
192: {|\langle {\rm 2ph}| \alpha_{20} | {\rm 1ph}\rangle|^2}.
193: \end{equation}
194: This construction is possible for $^{74}$Ge and $^{100}$Mo as
195: shown in Table II and the results are included in Table I. 
196: The information about the two-phonon state in $^{64}$Ni is uncertain 
197: but it is consistent with a harmonic vibration, which is what is 
198: assumed in Table I.
199: % The two-phonon coupling strength is also uncertain for $^{34}$S;
200: % the strength for the $0_2^+$ state is not known, and the $2_2^+$ state 
201: % shown in Table II was obtained by combining two $2^+$ states, at 3.304 
202: % and 4.115 MeV, with B-values of 4.0 and 3.2 W.u.
203: 
204: In general, it is useful to characterize the coupling matrix elements
205: between the $n$-phonon and $(n-1)$-phonon states in terms of an
206: effective $\beta$ value, $\beta_\lambda(n{\rm ph})$, which is defined by
207: \begin{equation}
208: \langle n{\rm ph}| \alpha_{\lambda0} | (n-1){\rm ph}\rangle =
209: \sqrt{n} \ \frac{\beta_\lambda(n{\rm ph})}{\sqrt{2\lambda+1}}.
210: \end{equation} 
211: The advantage of this representation is that the effective $\beta$-values
212: are identical in the harmonic oscillator model. Consistent with this definition
213: one can also define the effective B-value,
214: \begin{equation}
215: B(E\lambda,n{\rm ph}\rightarrow (n-1){\rm ph}) = 
216: \Bigl(\frac{3ZeR_C^\lambda}{4\pi}\Bigr)^2 \ 
217: \frac{n \ \beta_\lambda^2(n{\rm ph})}{2\lambda+1}.
218: \end{equation} 
219: 
220: When the structure of multi-phonon excitations deviates from the
221: harmonic oscillator model, it can become difficult to calculate matrix 
222: elements of a general nuclear interaction.  However, the calculation
223: is not that difficult if the interaction only contains terms that
224: are linear and quadratic in the deformation amplitudes, as in Eq. (1).
225: Matrix elements of the quadratic terms can be calculated by 
226: inserting a complete set $|n'{\rm ph}\rangle$ of intermediate states,
227: \begin{equation}
228: \langle n_2{\rm ph}| \alpha_{\lambda0}^2 | n_1{\rm ph}\rangle =
229: \sum_{n'} \
230: \langle n_2{\rm ph}| \alpha_{\lambda0} | n'{\rm ph}\rangle \
231: \langle n'{\rm ph}| \alpha_{\lambda0} | n_1{\rm ph}\rangle.
232: \end{equation}
233: The intermediate couplings in this expression that are not known will be 
234: estimated in the harmonic limit by extrapolation from nearby known states.
235: 
236: The information about octupole excitations is usually limited to the 
237: one-phonon excitation.  The octupole modes shown in Table I have 
238: been modeled as harmonic vibrations.
239: The basic two-phonon calculations, referred to as the PH-2 calculations,
240: will include all of the one- and two-phonon states shown in Table I, in 
241: addition to the mutual excitations of the one-phonon states. 
242: All calculations are furthermore restricted by a 7 MeV cutoff in 
243: excitation energy.  
244: 
245: \section{Systematics of comparison with data}
246: 
247: The measured fusion cross sections for $^{64}$Ni on the three targets: 
248: $^{64}$Ni \cite{nini}, $^{74}$Ge \cite{Beck}, and $^{100}$Mo \cite{nimo}, 
249: are compared to two sets of calculations in Fig. 1. The solid curves show
250: the basic two-phonon coupled-channels calculations (PH-2) described in 
251: the previous section, and the dashed curves are the results obtained in 
252: the no-coupling limit using the same ion-ion potential.
253: There is an essentially constant energy shift between the solid and dashed 
254: curves at low energies, and the energy shift is seen to increase as the 
255: target gets heavier and softer.
256: 
257: All calculations shown in Fig. 1 were based on the value $\Delta R$ = 0.1 fm 
258: in the ion-ion potential, Eq. (2).  This value was chosen because it provides 
259: the best fit to the $^{64}$Ni+$^{64}$Ni fusion data that are larger than 0.1 mb.
260: The best fit to all of the $^{64}$Ni+$^{64}$Ni data is actually quite poor,
261: with a $\chi^2/N$=10.  The reason is that the coupled-channels calculations 
262: cannot reproduce the fusion hindrance that occurs far below the Coulomb barrier 
263: \cite{nini}, so it is better to exclude that energy region from the analysis. 
264: 
265: The $\chi^2$ values per point for different calculations are summarized 
266: in Table III.  By comparing the values obtained in the one-phonon (PH-1) 
267: and two-phonon (PH-2) calculations it is seen that the mutual and two-phonon 
268: excitations play a very important role in improving the fit to the data. 
269: The improvement is mainly caused by the second-order (or quadratic) term 
270: in the nuclear interaction, Eq. (1). This has been demonstrated previously 
271: in Refs. \cite{eslan,Takiful}.
272: It is also seen that the PH-2 fit to the $^{64}$Ni+$^{74}$Ge data is not 
273: as good as the fit to the $^{64}$Ni+$^{64}$Ni data, and the fit to the 
274: $^{64}$Ni+$^{100}$Mo data is even worse. In the next section we shall
275: try to improve the fits by including more channels in the calculations.  
276: 
277: The third column shows the overall energy shift $\Delta E$ that is 
278: required to minimize the $\chi^2$ for a given calculation and data set.
279: It is seen that the shift depends on the channels that are included 
280: in the calculations.  The energy shift is effectively equivalent to 
281: changing the radius parameter $\Delta R$ of the ion-ion potential, 
282: Eq. (\ref{BWP}).
283: Thus one should realize that one cannot easily beforehand predict 
284: the best value of $\Delta R$, because the value that gives the best 
285: fit to the data depends on channels that are included in the calculation. 
286: This feature was recognized in Ref. \cite{Takipol} in a study of
287: the influence of couplings to high-lying states on fusion.
288: Another example is the two-phonon $3^-$ state in $^{64}$Ni, which was 
289: excluded from the calculations by the 7 MeV cutoff in excitation energy.
290: If this state and also the mutual excitations of the $3^-$ states in projectile
291: and target are included in the analysis of the $^{64}$Ni+$^{64}$Ni
292: fusion data above 0.1 mb, one obtains the same good fit as shown
293: in Table III but the required energy shift is now $\Delta E$ = 0.27 MeV. 
294: The same fit is achieved by reducing $\Delta R$ from 0.10 fm to 0.07 fm. 
295: 
296: \subsection{Logarithmic derivative}
297: 
298: One way to illustrate the behavior of the fusion cross section at low 
299: center-of-mass energies, $E_{c.m.}$, is to plot the logarithmic derivative 
300: of the energy weighted fusion cross section, which is defined as \cite{syst}
301: \begin{equation}
302: \label{logderiv}
303: L(E_{c.m.}) = \frac{1}{E_{c.m.}\sigma_f} \ 
304: \frac{d(E_{c.m.}\sigma_f)}{dE_{c.m.}}.
305: \end{equation}
306: The results for the three heavy-ion systems are shown in Fig. 2. 
307: The experimental values are seen to increase steeply with decreasing energy. 
308: The top lines in Fig. 2 show the logarithmic derivative for constant $S$ 
309: factor \cite{syst}, i.~e. $L_{CS}=\pi\eta/E_{c.m.}$, where $\eta$ is the 
310: Sommerfeld parameter. It is seen that the data for the $^{64}$Ni and 
311: $^{100}$Mo targets intersect the constant $S$ factor curves. 
312: The experimental $S$ factor, $S=E_{c.m.}\sigma_f\exp(2\pi\eta)$, will 
313: therefore exhibit a maximum, and the energy where that occurs has been used to 
314: characterize the onset of the low-energy fusion hindrance \cite{nini,nimo}. 
315: The data for the $^{74}$Ge target have not reached the constant 
316: $S$ factor limit but it is likely they will if measurements were performed 
317: at lower energies.
318: 
319: The logarithmic derivatives obtained from the no-coupling calculations 
320: rise steeply near the Coulomb barrier but level off at lower energies.  
321: The results for the PH-2 coupled-channels calculations 
322: show a similar behavior. They are just shifted to lower energies.
323: They exhibit some oscillations at the lowest energies. The magnitude 
324: of the oscillations depend on where the ingoing wave boundary conditions 
325: are imposed and on the strength of the imaginary optical potential.
326: However, the oscillations are not essential since they have not been
327: observed experimentally. The important point is that it has not yet 
328: been possible, within the coupled-channels approach, to develop a 
329: credible description that reproduces the data at the lowest energies.
330: 
331: \subsection{High-energy behavior}
332: 
333: The cross sections shown in Fig. 1 are plotted on a linear scale 
334: in Fig. 3. It is unfortunate that the measurements  do not reach 
335: larger cross sections, so it is difficult to assess the suppression 
336: of the data compared to the no-coupling limit, which was identified 
337: in Ref. \cite{newt1} for cross sections larger than 200 mb.
338: It is seen, however, that the coupled-channels calculations agree 
339: fairly well with the data points that are above the 200 mb limit,
340: and they are shifted to higher energies (i.e., they are suppressed) 
341: when compared to the no-coupling calculations. 
342: The suppression is mainly caused by the long-range Coulomb excitation
343: of the low-lying quadrupole states, and it is largest for the $^{74}$Ge 
344: and $^{100}$Mo targets, consistent with the fact that the quadrupole 
345: mode is particularly soft in these two nuclei.
346: 
347: Another way to illustrate the behavior of the fusion cross sections 
348: for the three systems is to plot the derivative of $E_{c.m.}\sigma_f$, 
349: which is shown in Fig. 4.  The behavior at energies far above the Coulomb 
350: barrier, $V_{CB}$, is often parametrized as 
351: \begin{equation}
352: \sigma_f=\pi R_f^2 \ \Bigl(1-\frac{V_{CB}}{E_{c.m.}}\Bigr).
353: \end{equation}
354: The derivative of $E_{c.m.}\sigma_f$ should therefore approach the constant 
355: value $\pi R_f^2$ at high energy. The curves shown in Fig. 4 do approach 
356: a constant value at high energies and the data are also consistent with that 
357: behavior. There is some uncertainty in the highest data point for the 
358: $^{64}$Ni+$^{100}$Mo system because the estimated contribution from 
359: fission is large. It is also seen that the data are suppressed at energies 
360: near the Coulomb barrier and enhanced at lower energies compared to the 
361: PH-2 calculation. There are similar but more modest discrepancies for 
362: the $^{64}$Ni+$^{74}$Ge system.
363: 
364: \section{Effects of multi-phonon excitations}
365: 
366: A simple way to expand the PH-2 calculations presented in the
367: previous section is to include all mutual excitations of the
368: states given in Table I up to three-phonon (2PH-3) or four-phonon
369: excitations (2PH-4). The results of the $\chi^2$ analysis using such 
370: calculations, again with a maximum excitation energy cutoff of 7 MeV, 
371: are shown in Table III for the $^{74}$Ge and $^{100}$Mo targets, and also 
372: for the $^{74}$Ge+$^{74}$Ge systems, which will be discussed below.
373: It is seen that the expanded calculations give a much better fit to the 
374: $^{64}$Ni+$^{74}$Ge data, whereas the improvements for the heavier 
375: systems are modest. 
376: 
377: In addition to the mutual excitations discussed above, one can
378: also explicitly include three- and four-phonon excitations in
379: the calculations. The problem is that the energy and transition 
380: strengths of such states are poorly known. The best known transitions
381: in heavy nuclei are the quadrupole transitions and below we estimate
382: the effect on fusion of couplings to an effective three-phonon 
383: quadrupole state.  There are many states at the three-phonon level 
384: but it is possible to lump them together into one effective 
385: three-phonon state as described in the Appendix.
386: 
387: \subsection{Calculations for $^{64}$Ni+$^{74}$Ge}
388: 
389: The quadrupole mode in $^{74}$Ge is rather soft and there 
390: are several even parity states at the three-phonon level.
391: Unfortunately, the knowledge about the couplings to these states 
392: is poor so it is not possible to construct an effective three-phonon
393: state using the method described in the Appendix.
394: It is therefore assumed in the following that the three-phonon 
395: quadrupole state is at 2 MeV and that the associated 
396: $\beta$-values are the same as for the two-phonon state, i.~e.,
397: $\beta_2^C$(3ph)=0.217 and $\beta_2^N$(3ph)=0.239.
398: 
399: The calculations that include the three-phonon quadrupole mode, 
400: in addition to all of the mutual excitations of the states shown 
401: in Table I up to the four-phonon level, are denoted by 3PH-4. 
402: These calculations are again restricted to excitation energies 
403: below a 7 MeV cutoff and consist of 35 coupled channels.  
404: They can be compared to the 2PH-4 calculations, which do not
405: include the three-phonon quadrupole mode in $^{74}$Ge. 
406: From table III it is concluded that the three-phonon state is not 
407: very important since both calculations require the same energy 
408: shift to fit the data and give essentially the same $\chi^2$.
409: The required energy shift of -0.57 MeV is equivalent 
410: to increasing the $\Delta R$ parameter from 0.10 to 0.16 fm.
411: 
412: The calculated 3PH-4 cross sections are shown by the solid curve in 
413: the top part of Fig. 5. The results of the PH-2 and 
414: PH-1 calculations and the no-coupling limit are also shown for 
415: comparison.  All calculations shown here were based on the parameter 
416: value $\Delta R$=0.16 fm, which minimizes the $\chi^2$ fit to the 
417: data in the 3PH-4 calculations. It is seen that the 3PH-4 and
418: the PH-2 calculations do not differ much, except at the lowest energies.
419: In other words, multi-phonon excitations (beyond PH-2) do not play
420: a very dramatic role in the calculation of the $^{64}$Ni+$^{74}$Ge
421: fusion cross section; they just provide a fine-tuning that produces 
422: a better $\chi^2$ fit to the data.
423: 
424: The logarithmic derivatives of the cross sections are
425: shown in the bottom part of Fig. 5. 
426: The shape of the calculated $L(E_{c.m.})$ clearly improves
427: in comparison with the low-energy data as more channels are 
428: included in the calculations. 
429: Some discrepancy is beginning to develop at the lowest energy 
430: where the calculated $L(E_{c.m.})$ saturates, whereas 
431: the measured values keep growing with decreasing energy.
432: This behavior is consistent with the systematic trend of the 
433: fusion hindrance phenomenon at extreme subbarrier \cite{nimo}, 
434: which was observed in Fig. 2 for the other two systems.
435: 
436: \subsection{Calculations for $^{64}$Ni+$^{100}$Mo}
437: 
438: The experimental structure information about the states associated with the
439: three-phonon quadrupole excitation in $^{100}$Mo is shown in Table IV.
440: The excitation energies and most of the couplings to two-phonon states 
441: are known, and the two that are not known can be estimated in the 
442: harmonic oscillator model (the values in parenthesis). 
443: In that model one has the sum rule
444: \begin{equation}
445: \sum_{I_2} B(E2,I_3\rightarrow I_2) = 3 \ B(E2, 2\rightarrow 0),
446: \end{equation}
447: i.~e., the sum is three times the B-value for the one-phonon 
448: excitation.  That would give a sum of 112 W.u., according to Table I. 
449: It is seen that the sums in Table IV for different values of the spin 
450: $I_3$ are less than the harmonic limit. 
451: 
452: We can now use the expressions derived in the Appendix to estimate
453: the properties of the effective three-phonon state. The expression
454: (\ref{alf3}) is based on amplitudes, whereas the B-values are proportional 
455: to the square of these amplitudes. To proceed we assume that the 
456: amplitudes are positive and determined by the square root of the 
457: B-values. This assumption is correct for the harmonic oscillator.
458: If we use all of the transition strengths shown in Table IV,
459: including the two harmonic estimates of the two unknown strengths
460: (indicated by the `?'), we obtain $\beta_2^C({\rm 3ph})$ = 0.182 
461: and a three-phonon excitation energy of 1.65 MeV.  
462: The nuclear coupling is set 10\% higher, i.~e. $\beta_2^N$(3ph)=0.2.
463: These are the values that will be used in the 3PH-3 and 3PH-4
464: calculations discussed below. 
465: 
466: The estimated three-phonon coupling strength is weaker than the 
467: harmonic limit (which gives $\beta_2^C=0.231$) but the excitation 
468: energy is not much different from the harmonic value (1.61 MeV). 
469: Another limit is to include only the known transition strengths 
470: in Table IV, and exclude the harmonic estimates of the
471: two unknown strengths. That gives the value $\beta_2^C({\rm 3ph})$ 
472: = 0.161 and an excitation energy of 1.70 MeV. 
473: This shows that the three-phonon excitation energy is quite 
474: accurately determined, whereas the $\beta$-value may have 
475: an uncertainty of 10\%,
476: 
477: The results of the $\chi^2$ analysis,
478: which was based on the radius parameter value $\Delta R$ = 0.10 fm,
479: are shown in Table III. The best fit is obtained with the 3PH-3 calculation 
480: but the fit is still poor. When the analysis is restricted to cross sections 
481: that are larger than 0.1 mb, the $\chi^2/N$ is reduced from 25 to 15.
482: The poor fit to the data is therefore not entirely due to the fusion 
483: hindrance phenomenon, which occurs at smaller cross sections.
484: 
485: Inspecting the fusion data for $^{64}$Ni+$^{100}$Mo shown in Fig. 1 
486: it appears that the fusion hindrance sets in at cross sections that are
487: smaller than 1 $\mu$b.  To fit the data above 1 $\mu$b with the 3PH-3 
488: calculation requires an energy shift of 0.7 MeV, or a new radius
489: parameter of $\Delta R$ = 0.04 fm. To make a reasonable comparison 
490: with the data, all calculations shown in Fig. 6 have therefore been 
491: based on this value for the radius parameter.
492: 
493: In the top panel of Fig. 6 it is seen that the 3PH-3 calculation agrees 
494: with the largest cross section and the 0.92 mb cross section measured 
495: at 127.5 MeV, but it is too high in between.
496: The good agreement at the highest energy could be misleading because
497: 50\% of the fusion cross section that is shown (solid point) is an
498: estimated contribution from fission. The measured evaporation residue
499: cross sections are shown by the open circles. If the correction for 
500: fission were smaller, then there would be a general suppression of 
501: the data at energies above the Coulomb barrier, in agreement with 
502: systematics \cite{newt1,newt2}.  
503: 
504: The discrepancy between calculations and measurements at low 
505: energies is emphasized by comparing the logarithmic derivatives, 
506: which are shown in the bottom part of Fig. 6.
507: The discrepancy is large, and the calculated curves fail to
508: reproduce the value at the lowest energy, where the fusion hindrance 
509: phenomenon sets in.  Let us finally show some results for $^{74}$Ge+$^{74}$Ge 
510: in order see how the discrepancies evolve with mass asymmetry.
511: 
512: \subsection{Results for $^{74}$Ge+$^{74}$Ge}
513: 
514: The results of calculations for the $^{74}$Ge+$^{74}$Ge system,
515: using the radius parameter value $\Delta R$ = 0.10 fm, are compared 
516: with the data \cite{Begege} in Fig. 7.
517: The data are the measured evaporation residue cross sections
518: whereas the fission cross sections were estimated to be small,
519: about 15 mb at the highest energy.
520: The three-phonon quadrupole state used in the 3PH-4 calculation is
521: the same as used in Sect. IV.A, whereas the 2PH-4 calculation
522: does not include that state. By comparing the two calculations
523: it is seen that couplings to the three-phonon quadrupole state have 
524: a significant influence and improves the shape of the calculated 
525: cross section in comparison with the data. Thus it reduces the 
526: cross section at higher energies and enhances it at energies below 
527: the Coulomb barrier. This is what is needed to improve the fit
528: to the data but the discrepancy with the data remains large, 
529: with a $\chi^2/N$=17. The demonstrated sensitivity to the three-phonon 
530: quadrupole state indicates that the calculation has not converged.  
531: 
532: The logarithmic derivatives are shown in the bottom part of Fig. 7.
533: The two coupled-channels calculations exhibit oscillations 
534: that are out of phase at low energy and the 2PH-4 calculation 
535: has a bump near 120 MeV. These differences should not be taken too
536: seriously because the calculations have not converged with respect 
537: to multi-phonon excitations, as explained above. 
538: The discrepancies with the data in Fig. 7 are similar to 
539: those seen in Fig. 6 for the fusion of $^{64}$Ni+$^{100}$Mo. 
540: The measured fusion cross sections are suppressed at high energies
541: compared to the calculations, in agreement with systematics 
542: \cite{newt1,newt2}, and they are enhanced below the Coulomb barrier, 
543: where they fall off at a slower pace when compared to the calculations. 
544: 
545: \subsection{Further improvements}
546: 
547: In order to improve the calculations of the fusion of heavy, soft nuclei, 
548: one would have to include excitations to higher multi-phonon states. 
549: That would create several problems. One problem is that the structure 
550: of such states is often poorly known.
551: Another problem is that the nuclear interaction, Eq. (1), only 
552: includes terms up to second order in the deformation amplitudes, 
553: whereas matrix elements associated with multi-phonon states 
554: would be sensitive to higher-order terms.
555: In the harmonic oscillator model one can calculate the nuclear 
556: coupling matrix elements to all orders in the deformation 
557: amplitudes \cite{eslan,Takiful}. Unfortunately, the connection 
558: to the actual structure of the reacting nuclei may be poor in such 
559: calculations. The most practical solution to these problems would 
560: be to develop and apply more realistic structure models, such as 
561: those presented in Ref. \cite{Takianh}.
562: 
563: \section{Conclusions}
564: 
565: The calculated fusion cross sections that were presented and compared
566: to measurements illustrate nicely how multi-phonon excitations play an 
567: increasingly important role as the fusing nuclei become softer and heavier. 
568: The basic two-phonon calculations reproduce quite well the data for 
569: the lightest system $^{64}$Ni+$^{64}$Ni (except at extreme subbarrier 
570: energies) but it is necessary to include up to three- or four-phonon 
571: excitations in order to reproduce the $^{64}$Ni+$^{74}$Ge fusion data. 
572: In the analysis of the data for the two heavy systems, 
573: $^{64}$Ni+$^{100}$Mo and $^{74}$Ge+$^{74}$Ge, it was not possible 
574: to achieve a good fit by including up to four-phonon excitations.  
575: The subbarrier fusion data were enhanced and the data above the Coulomb 
576: barrier were suppressed compared to the most complete calculations. 
577: It is difficult to ascertain which of these two discrepancies is the most 
578: dominant because the radius parameter of the ion-ion potential, which 
579: determines the energy scale of the calculated fusion cross sections, 
580: cannot be predicted accurately.
581: 
582: The observed suppression of the data above the Coulomb barrier compared 
583: to the no-coupling limit is consistent with the results of the systematic 
584: studies by Newton et al. \cite{newt1,newt2}. The fusion cross sections 
585: obtained in the coupled-channels calculations are also suppressed 
586: compared to the no-coupling limit.
587: This is caused by the long-range Coulomb excitation
588: which pushes the surfaces of the reacting nuclei away from each other 
589: whereby the attractive nuclear force is reduced.  However, the calculated 
590: suppression is apparently not large enough to explain the data. 
591: It has been suggested that the remaining discrepancy could be due to 
592: deep inelastic reactions \cite{newt2}.
593: 
594: The enhancement of the data at energies below the Coulomb barrier
595: (ignoring for a moment the hindrance phenomenon at extreme subbarrier energies) 
596: indicates that additional channels or couplings need to be included
597: in the coupled-channels calculations. 
598: One possibility is transfer reactions but they should play a minor 
599: role in the fusion of the symmetric $^{74}$Ge+$^{74}$Ge system. 
600: The most natural explanation is the limitations of the nuclear 
601: interaction which was expanded up to second order in the deformation 
602: parameters. This approximation works very well in calculations 
603: of fusion cross sections for lighter and stiff systems,
604: but it becomes unrealistic for heavy and soft systems, 
605: where the calculations become sensitive to multi-phonon excitations 
606: and therefore to the nuclear interaction at large deformation amplitudes
607: \cite{Takiful}.
608: 
609: A serious complication and uncertainty in the calculation of fusion 
610: cross sections for heavy and soft nuclei is the sensitivity to poorly 
611: known multi-phonon states.  
612: One example is the two-phonon octupole state. Fortunately, this 
613: state did not play a very crucial role in the calculations that 
614: were presented for $^{64}$Ni+$^{64}$Ni. 
615: The main effect of couplings to this state was an overall shift in 
616: energy, whereas the shape of the cross section was not affected. 
617: The knowledge of quadrupole excitations is much better.
618: Thus it was demonstrated that it is possible to construct effective 
619: two- and three-phonon states in $^{100}$Mo from detailed structure 
620: information.  Such a construction may become more difficult 
621: at the four-phonon level.  
622: 
623: In addition to improving the structure input for multi-phonon excitations
624: and the nuclear interaction at large deformations, one should also seek 
625: an explanation for what causes the fusion hindrance in measurements at 
626: extreme subbarrier energies. That could also be related to the 
627: parametrization of the ion-ion potential, as suggested by Dasso 
628: and Pollarolo \cite{Dasso}.
629: 
630: \begin{acknowledgments} 
631: This work was supported by the U.S. Department of Energy, 
632: Office of Nuclear Physics, under Contract No. W-31-109-ENG-38.
633: \end{acknowledgments}
634: 
635: \appendix*
636: \section{Two- and three-phonon excitations}
637: 
638: The number of states at the two- and three-phonon level of
639: quadrupole excitations is so large that it is convenient to lump 
640: them together into effective two- and three-phonon states. 
641: The procedure for doing that in the rotating frame approximation was 
642: discussed in Ref. \cite{alge} for two-phonon excitations and below 
643: it is shown how it can be generalized to three-phonon level.
644: 
645: Let us include all couplings that are linear and quadratic in the 
646: deformation amplitudes and denote the associated radial form
647: factors by $F_1(r)$ and $F_2(r)$. 
648: The radial wave functions for the $n$-phonon state with spin $I$ 
649: is denoted by $\psi_n(I)$, and the coupling of the $(n,I)$ state 
650: to the $(n-1,I_{n-1})$ state is
651: \begin{equation}
652: \alpha_n(I_n,I_{n-1}) =
653: \langle I_{n-1} 0 \ 20| I_n0\rangle
654: \frac{\langle nI||\alpha_2|| (n-1)I_{n-1}\rangle} {\sqrt{2I_n+1}},
655: \end{equation}
656: according to Eq. (4).  The notation can be simplified for the one- 
657: and two-phonon couplings by defining $\alpha_1=\alpha_1(2,0)$ 
658: and $\alpha_2(I)=\alpha_2(I,2)$.
659: With these definitions, the coupled equations have the following 
660: form in the rotating frame approximation 
661: \begin{equation}
662: \label{cc0}
663: (H-E) \psi_0 = - F_1 \ \alpha_1 \ \psi_1 
664: - F_2 \sum_I \alpha_1 \alpha_2(I) \ \psi_2(I),
665: \end{equation}
666: \begin{equation}
667: \label{cc1}
668: (H - E_1) \psi_1 = - F_1 \alpha_1 \ \psi_0 - 
669: F_1 \sum_{I_2} \alpha_2(I_2) \ \psi_2(I_2) - F_2
670: \sum_{I_2,I_3} \alpha_2(I_2) \alpha_3(I_3,I_2) \ \psi_3(I_3),
671: \end{equation}
672: \begin{equation}
673: \label{cc2}
674: (H - E_2(I_2)) \psi_2(I_2) =  
675: - F_2 \alpha_2(I_2) \alpha_1 \ \psi_0 
676: - F_1 \alpha_2(I_2) \ \psi_1 
677: - F_1 \sum_{I_3} \alpha_3(I_3,I_2) \ \psi_3(I_3),
678: \end{equation}
679: \begin{equation}
680: \label{cc3}
681: (H - E_3(I_3)) \psi_3(I_3) = 
682: - F_1 \sum_{I_2} \alpha_3(I_3,I_2) \ \psi_2(I_2)
683: - F_2 \sum_{I_2} \alpha_3(I_3,I_2) \alpha_2(I_2) \ \psi_1,
684: \end{equation}
685: where $H$ is the scattering Hamiltonian in the rotating frame,
686: which is assumed to be the same for all channels. 
687: The available energy in the channel $(n,I)$ is denoted by 
688: $E_n(I)=E-\epsilon_n(I)$, where $\epsilon_n(I)$ is the excitation energy. 
689: 
690: The structure of the above equations is actually quite simple.
691: An effective two-phonon state, with coupling strength $\alpha_2$
692: and scattering wave function $\psi_2$, can be introduced by 
693: the substitution
694: \begin{equation}
695: \label{ph2}
696: \alpha_2 \psi_2 =  \sum_I \alpha_2(I) \ \psi_2(I), 
697:  \ \ {\rm where} \ \ \ \alpha_2^2 = \sum_I \alpha_2(I)^2.    
698: \end{equation}
699: A similar substitution can be made for the three-phonon state,
700: \begin{equation}
701: \label{ph3}
702: \alpha_2\alpha_3 \ \psi_3 = 
703: \sum_{I_2,I_3} 
704: \alpha_2(I_2) \alpha_3(I_3,I_2) \ \psi_3(I_3).
705: \end{equation}
706: where the three-phonon coupling strength $\alpha_3$ is determined by
707: \begin{equation}
708: \label{alf3}
709: (\alpha_2\alpha_3)^2 = \sum_{I_3} 
710: \Bigl[\sum_{I_2} \alpha_2(I_2) \alpha_3(I_3,I_2)\Bigr]^2.
711: \end{equation}
712: These substitutions simplifies the coupled equations 
713: (\ref{cc0}-\ref{cc3}).
714: The first two become
715: \begin{equation}
716: \label{cef0}
717: (H-E) \psi_0 = - F_1 \ \alpha_1 \ \psi_1 
718: - F_2 \alpha_1 \alpha_2 \ \psi_2,
719: \end{equation}
720: \begin{equation}
721: \label{cef1}
722: (H - E_1) \psi_1 = - F_1 \alpha_1 \ \psi_0 - 
723: F_1 \alpha_2 \ \psi_2 - F_2
724: \alpha_2\alpha_3 \ \psi_3,
725: \end{equation}
726: which is just the form we want for a zero- and one-phonon state that 
727: are coupled to each other and to the two- and three-phonon states.
728: 
729: The Eq. (\ref{cc2}) can be brought into a similar
730: form if we assume that the two-phonon energies
731: are almost degenerate, $E_2(I)\approx E_2$.
732: We can then multiply Eq. (\ref{cc2}) by $\alpha_2^{-1}\alpha_2(I_2)$,
733: sum over $I_2$, and use the substitutions (\ref{ph2}) and (\ref{ph3}) 
734: to obtain
735: \begin{equation}
736: \label{cef2}
737: (H - E_2) \psi_2 =  
738: - F_2 \alpha_2 \alpha_1 \ \psi_0 
739: - F_1 \alpha_2 \ \psi_1 
740: - F_1 \alpha_3 \ \psi_3.
741: \end{equation}
742: 
743: The Eq. (\ref{cc3}) can be dealt with in a similar way assuming a
744: near degeneracy of the three-phonon states, $E_3(I)\approx E_3$.
745: Thus we can multiply Eq. (\ref{cc3}) by
746: $(\alpha_2\alpha_3)^{-1}\alpha_2(I_2)\alpha_3(I_3,I_2)$, sum over 
747: $I_2$ and $I_3$, and use Eqs. (\ref{ph3}) and (\ref{alf3}) to obtain
748: \begin{equation}
749: \label{cef3}
750: (H - E_3) \psi_3 = 
751: - F_1 \alpha_3 \ \psi_2
752: - F_2 \alpha_3\alpha_2 \ \psi_1.
753: \end{equation}
754: Actually, to derive the first term on the right hand side of Eq. 
755: (\ref{cef3}) one also needs the assumption 
756: $\psi_2(I_2)$ = $\alpha_2^{-1}\alpha_2(I_2) \psi_2$, which 
757: is not unreasonable because it is consistent with Eq. (\ref{ph2}) 
758: and it is exact for $\alpha_3=0$, as can be seen from Eqs. (\ref{cc2}) 
759: and (\ref{cef2}).
760: That completes the derivation.  If the states at the two- or
761: three-phonon level are not quite degenerate one can construct the 
762: effective two- and three-phonon excitation energies from the energy 
763: weighted sums
764: \begin{equation}
765: \epsilon_2 = \frac{1}{\alpha_2^2} \
766: \sum_I |\alpha_2(I)|^2 \ \epsilon_2(I),
767: \end{equation}
768: \begin{equation}
769: \epsilon_3 = \frac{1}{(\alpha_2\alpha_3)^2}  
770: \sum_{I_3} \Bigl[\sum_{I_2} \alpha_2(I_2) \alpha_3(I_3,I_2)\Bigr]^2 \ 
771: \epsilon_3(I_3).
772: \end{equation}
773: % In the harmonic oscillator model one obtains the familiar
774: % expressions: $\alpha_2=\sqrt{2}\alpha_1$ and $\alpha_3=\sqrt{3}\alpha_1$ 
775: % in terms of the one-phonon coupling strength $\alpha_1$.
776: 
777: \begin{thebibliography}{99}
778: \bibitem{baha} A. B. Balantekin and N. Takigawa,  Rev. of Mod. Phys. {\bf 70}, 77 (1998).
779: \bibitem{newt1} J. O. Newton, R. D. Butt, M. Dasgupta, D. J. Hinde, I. I. Gontchar,
780: C. R. Morton, and K. Hagino, Phys. Lett. B {\bf 586}, 219 (2004).
781: \bibitem{newt2} J. O. Newton, R. D. Butt, M. Dasgupta, D. J. Hinde, I. I. Gontchar,
782: C. R. Morton, and K. Hagino, Phys. Rev. C {\bf 70}, 024605 (2004).
783: \bibitem{syst} C. L. Jiang, H. Esbensen, B. B. Back, R. V. F. Janssens, and
784: K. E. Rehm, Phys. Rev. C {\bf 69}, 014604 (2004).
785: \bibitem{eslan} H. Esbensen and S. Landowne, Phys. Rev. C {\bf 35}, 2090 (1987).
786: \bibitem{Takiful} K. Hagino, N. Takigawa, M. Dasgupta, D. J. Hinde, and
787: J. R. Leigh, Phys. Rev. C {\bf 55}, 276 (1997). 
788: \bibitem{alge} H. Esbensen, Phys. Rev. C {\bf 68}, 034604 (2003).
789: \bibitem{Takipol} K. Hagino, N. Takigawa, M. Dasgupta, D. J. Hinde, and
790: J. R. Leigh, Phys. Rev. Lett. {\bf 79}, 2014 (1997). 
791: \bibitem{nini} C. L. Jiang {\it et al}., Phys. Rev. Lett. {\bf 93}, 012701 (2004).
792: \bibitem{Beck} M. Beckerman, M. Salomaa, A. Sperduto, J. D. Molitoris, and
793: A. DiRienzo, Phys. Rev. C {\bf 25}, 837 (1982).
794: \bibitem{nimo} C. L. Jiang {\it et al}., Phys. Rev. C {\bf 71}, 044613 (2005).
795: \bibitem{Begege} M. Beckerman, M. K. Salomaa, J. Wiggins, R. Rohe,
796: Phys. Rev. C {\bf 28}, 1963 (1983).
797: \bibitem{BW} R. A. Broglia and A. Winther, {\it Heavy Ion Reactions},
798: Lecture Notes (Frontier in Physics 1991).
799: \bibitem{NNDC} Evaluated Nuclear Structure Data Files, National Nuclear Data
800: Center, Brookhaven National Laboratory; http://www.nndc.bnl.gov/
801: \bibitem{Lecomte} R. Lecomte, M. Irshad, S. Landsberger, G. Kajrys, P. Paradis, 
802: and S. Monaro, Phys. Rev. C {\bf 22}, 2420 (1980).
803: \bibitem{Spear} R. H. Spear, At. Data and Nucl. Data. Tables, {\bf 42}, 55 (1989). 
804: \bibitem{Flem} F. Videbaek, P. R. Christensen, Ole Hansen, and K. Ulbak,
805: Nucl. Phys. A {\bf 256}, 301 (1976).
806: \bibitem{Takianh} K. Hagino, S. Kuyucak, and N. Takigawa, Phys. Rev. C {\bf 57},
807: 1349 (1998).   
808: \bibitem{Dasso} C. H. Dasso and G. Pollarolo, Phys. Rev. C {\bf 68}, 054604 (2003). 
809: \end{thebibliography}
810: 
811: \begin{table}
812: \caption{Excitation energies and coupling strengths used in the calculations.
813: The values for $^{64}$Ni are the same as used in \cite{nini}.
814: The B-values for $^{74}$Ge are from Refs. \cite{Lecomte,Spear} and from
815: Ref. \cite{NNDC} for $^{100}$Mo.
816: The effective two-phonon quadrupole states are determined in Table II,
817: and the two-phonon octupole states are estimated in the harmonic 
818: oscillator model.}
819: \begin{ruledtabular}
820: \begin{tabular} {|c|c|c|c|c|c|}
821: Nucleus &
822: $\lambda^\pi$ &  $E_x$ (MeV) & 
823: B(E$\lambda$) (W.u.) & \ $\beta_\lambda^C$ & $\beta_\lambda^N$ \\
824: \colrule
825: $^{64}$Ni  & $2^+$   & 1.346 & 8.6    & 0.165 & 0.185  \\
826:         & 2ph($2^+$) & 2.692 & (17.2) & 0.165 & 0.185  \\
827:  Z=28      & $3^-$   & 3.560 & 12     & 0.193 & 0.200  \\
828: \colrule
829: $^{74}$Ge  & $2^+$   & 0.596 &   33 & 0.285 & 0.314  \\
830:         & 2ph($2^+$) & 1.362 &   38 & 0.217 & 0.239  \\
831:   Z=32     & $3^-$   & 2.536 &  8.8 & 0.145 & 0.160  \\
832:         & 2ph($3^-$) & 5.072 & (17.6) & 0.145 & 0.160 \\
833: \colrule
834: $^{100}$Mo & $2^+$   & 0.536 &  37.4 & 0.231 & 0.254  \\
835:         & 2ph($2^+$) & 1.002 &  68  & 0.222 & 0.244  \\ 
836:   Z=42     & $3^-$   & 1.908 &  35  & 0.220 & 0.242  \\
837:         & 2ph($3^-$) & 3.816 & (70) & 0.220 & 0.242  \\
838: \end{tabular}
839: \end{ruledtabular}
840: \end{table}
841: 
842: \begin{table}
843: \caption{Energies and reduced transition probabilities (in W.u.) of the $0^+$, 
844: $2^+$, and $4^+$ states associated with a two-phonon quadrupole excitation. 
845: The values for 
846: % $^{64}$Ni, and 
847: $^{100}$Mo are from \cite{NNDC}, and the
848: values for $^{74}$Ge are from \cite{Lecomte}. The last two columns shows
849: the energy and coupling strength of the effective two-phonon state.}
850: \begin{ruledtabular}
851: \begin{tabular} {|c|c|c|c|c|c|c|}
852: Nucleus &  States:   & $0_2^+$ & $2_2^+$ & $4_1^+$ & 
853: Eff. 2ph & $\beta_2$(2ph) \\ 
854: % \colrule
855: % $^{64}$Ni & $E_I$ (MeV):        & 2.867 & 2.277 & 2.610 & 2.87-2.75 &  -    \\
856: %      & B(E2,I$\rightarrow 2$): &  110  &   -   & $<$37 & 22-41 & 0.177-0.242  \\
857: \colrule
858: $^{74}$Ge & $E_I$ (MeV):        & 1.483 & 1.204 & 1.464 & 1.362 &  -    \\
859:       & B(E2,I$\rightarrow 2$): & $<$22  &  54.2 & 36.1  & $<$38.5 & $<$0.217  \\
860: \colrule
861: $^{100}$Mo & $E_I$ (MeV):           & 0.695 & 1.064 & 1.136 & 1.002 &  -    \\
862:            & B(E2,I$\rightarrow$2$^+$): &   92  &   51  &   69  &  68   & 0.222  \\
863: \end{tabular}
864: \end{ruledtabular}
865: \end{table}
866: 
867: \begin{table}
868: \caption{Analysis of fusion data for different heavy-ion systems
869: using the nuclear interaction, Eq. (1) and (2), with $\Delta R$=0.10 fm,
870: and a 7.0 MeV cutoff in excitation energy. } 
871: \begin{ruledtabular}
872: \begin{tabular} {|c|c|c|c|c|}
873: System  &  Calculation & $\Delta E$ (MeV) & $\chi^2/N$ & Data Ref. \\
874: \colrule
875: $^{64}$Ni+$^{64}$Ni & PH-2 &  0.9 &  10  & all data \cite{nini} \\
876:                     & PH-2 & 0.03 &  0.5 & $\sigma_f>$0.1 mb \\
877: \colrule
878: $^{64}$Ni+$^{74}$Ge  & PH-1  & -1.53  & 15.6 & all data \cite{Beck} \\
879:            & PH-2  & -0.62  & 6.4  &    \\
880:            & 2PH-3 & -0.56  & 3.9  &    \\
881:            & 2PH-4 & -0.62  & 3.1  &    \\
882:            & 3PH-4 & -0.57  & 2.6  &    \\
883: \colrule
884: $^{64}$Ni+$^{100}$Mo & PH-1   &  0.18 & 44   & all data \cite{nimo} \\
885:            & PH-2   &  0.   & 33   &          \\
886:            & 2PH-3  &  0.35 & 27   &          \\
887:            & 3PH-3  &  0.61 & 25   &          \\
888:            & 2PH-4  &  0.13 & 31   &          \\
889:            & 3PH-4  &  0.42 & 34   &          \\
890: \colrule
891: $^{64}$Ni+$^{100}$Mo & PH-1  &  0.20  &  38   & $\sigma_f>0.1$ mb \\
892:            & PH-2  &  0.30  &  24   & \cite{nimo} \\
893:            & 2PH-3 &  0.70  &  18   &              \\
894:            & 3PH-3 &  0.94  &  15   &              \\
895:            & 2PH-4 &  0.56  &  25   &              \\
896:            & 3PH-4 &  0.85  &  27   &              \\
897: \colrule
898: $^{74}$Ge+$^{74}$Ge & PH-1  & +4.50  & 30  & all data \cite{Begege} \\
899:                     & PH-2  & -0.50  & 25  &     \\
900:                     & 2PH-4 & +0.12  & 26  &     \\
901:                     & 3PH-4 & +0.20  & 17  &     \\
902: \end{tabular}
903: \end{ruledtabular}
904: \end{table}
905: 
906: \begin{table}
907: \caption{Three-phonon quadrupole excitation in $^{100}$Mo. 
908: The B-values are shown (in W.u.) for the known E2 transitions, 
909: from the ($I_3$ = 0, 2, 4, 6) three-phonon states to the
910: ($I_2$ = 0, 2, 4) two-phonon states \cite{NNDC}.
911: Unknown values are indicated by '?'.
912: The values in parenthesis were obtained for a harmonic vibration.
913: The second last row shows the sum of the B-values for each $I_3$ spin state.
914: The last row shows the excitation energies of the states associated 
915: with the three-phonon excitation.}
916: \begin{ruledtabular}
917: \begin{tabular} {|c|c|c|c|c|}
918: $B(E2,I_3\rightarrow I_2)$ &  $I_3$ =0  &  2  &  4   &   6   \\
919: \colrule
920:  $I_2$ = 0 &   ---   &  14 (52) &    ---    & ---  \\
921:     2      & ? (112) &  ?  (21) &  30 (59)  &  --- \\
922:     4      &   ---   &  36 (38) &  28 (53)  &  94 (112) \\
923: \colrule
924:   Sum =    & ? (112) & $>$ 50 (112) &  58 (112) &  94 (112) \\
925: \colrule
926: $E_x(I_3)$ (MeV) & 1.5046 & 1.4639 & 1.7715 & 1.8469 \\
927: \end{tabular}
928: \end{ruledtabular}
929: \end{table}
930: 
931: % \end{document}
932: 
933: \begin{figure}
934: \includegraphics[width = 12cm]{nibf1.eps}
935: \caption{\label{alf} Fusion cross sections for the three systems
936: $^{64}$Ni+$^{64}$Ni \cite{nini}, $^{64}$Ni+$^{74}$Ge \cite{Beck}, and
937: $^{64}$Ni+$^{100}$Mo \cite{nimo} as functions of the center-of-mass
938: energy $E_{c.m.}$. Note that the results for $^{64}$Ni+$^{100}$Mo have 
939: been shifted by -10 MeV.
940: The dashed curves show the no-coupling limit, and the solid curves are 
941: the results of the PH-2 coupled-channels calculations described in the text.
942: All calculations were based on the radius parameter $\Delta R$ = 0.10 fm.}
943: %, with up to two-phonon excitations.}
944: \end{figure}
945: 
946: \begin{figure}
947: \includegraphics[width =12cm]{nibf2.eps}
948: \caption{\label{al1d} Logarithmic derivatives of the fusion cross sections 
949: shown in Fig. 1. The top lines show the constant {\it S} factor limit.} 
950: \end{figure}
951: 
952: \begin{figure}
953: \includegraphics[width =12cm]{nibf3.eps}
954: \caption{\label{allf} Linear plot of the fusion cross sections shown in Fig. 1.}
955: \end{figure}
956: 
957: \begin{figure}
958: \includegraphics[width =12cm]{nibf4.eps}
959: \caption{\label{aldf} Derivative of $E_{c.m.}\sigma_f$, where the fusion cross 
960: sections $\sigma_f$ are shown in Fig. 1.}
961: \end{figure}
962: 
963: \begin{figure}
964: \includegraphics[width = 12cm]{nibf5.eps}
965: \caption{\label{nigeco2} Fusion cross sections for $^{64}$Ni+$^{74}$Ge 
966: \cite{Beck} (top panel) and the associated logarithmic derivatives
967: (bottom panel). The dashed curves show the no-coupling limit (NOC),
968: the one-phonon (PH-1), and the basic two-phonon (PH-2) calculations.
969: The solid curves show the results of the 3PH-4 calculations.
970: All calculations used the radius parameter $\Delta R$=0.16 fm.
971: The upper line in the bottom panel is the result for constant $S$ factor.}
972: \end{figure}
973: 
974: \begin{figure}
975: \includegraphics[width = 12cm]{nibf6.eps}
976: \caption{\label{nimoco2} Fusion cross sections for $^{64}$Ni+$^{100}$Mo 
977: (top panel) and the associated logarithmic derivatives (bottom panel). 
978: The open circles are the measured evaporation cross section; the solid 
979: points include an estimated fission cross section \cite{nimo}.
980: The dashed curves show the no-coupling limit (NOC), the one-phonon 
981: (PH-1), and the basic two-phonon (PH-2) calculations.
982: The solid curves are the results of the 3PH-3 calculation.
983: All calculations used the radius parameter $\Delta R$ = 0.04 fm.
984: The upper line in the bottom panel is the result for constant $S$ factor.}
985: \end{figure}
986: 
987: \begin{figure}
988: \includegraphics[width = 12cm]{nibf7.eps}
989: \caption{\label{gegeco2} Fusion cross sections for $^{74}$Ge+$^{74}$Ge 
990: \cite{Beck} (top panel) and the associated logarithmic derivatives
991: (bottom panel). 
992: The upper line in the bottom panel is the result for constant $S$ factor.
993: The dashed curves show the no-coupling limit (NOC), and the 2PH-4
994: calculations. The solid curves show the result of the 3PH-4 calculation.
995: All calculations used the radius parameter $\Delta R$ = 0.10 fm. 
996: The upper line in the bottom panel is the result for constant $S$ factor.}
997: \end{figure}
998: 
999: \end{document}
1000: