1: \documentclass[12pt]{report}
2: \thispagestyle{empty}
3: \usepackage[dvips]{graphicx}
4: \oddsidemargin=-.05in
5: \textwidth=6.5in
6: \topmargin=-.265in
7: \textheight=9in
8: \newcommand{\spone}{0.9}
9: \newcommand{\sphalf}{1.15}
10: \newcommand{\sptwo}{1.4}
11: \newcommand{\spthree}{2.4}
12: \newcommand{\singlespace}{\edef\baselinestretch{\spone}\Large\normalsize}
13: \newcommand{\kimspace}{\edef\baselinestretch{\sphalf}\Large\normalsize}
14: \newcommand{\doublespace}{\edef\baselinestretch{\sptwo}\Large\normalsize}
15: \newcommand{\threespace}{\edef\baselinestretch{\spthree}\Large\normalsize}
16: \begin{document}
17: \doublespace
18: \begin{center}
19: {\bf Hydrodynamic Modes in a Trapped Strongly Interacting Fermi Gases of Atoms }\\
20: \renewcommand\thefootnote{\fnsymbol{footnote}}
21: {Yeong E. Kim \footnote{ e-mail:yekim$@$physics.purdue.edu} and
22: Alexander L. Zubarev\footnote{ e-mail: zubareva$@$physics.purdue.edu}}\\
23: Purdue Nuclear and Many-Body Theory Group (PNMBTG)\\
24: Department of Physics, Purdue University\\
25: West Lafayette, Indiana 47907\\
26: \end{center}
27:
28: \begin{quote}
29: The zero-temperature properties of a dilute two-component Fermi gas in the
30: BCS-BEC crossover are investigated. On the basis of a generalization of the
31: variational Schwinger method,
32: we construct approximate semi-analytical formulae for
33: collective
34: frequencies of the radial and the axial breathing modes of the Fermi gas
35: under harmonic confinement in the framework of the
36: hydrodynamic theory.
37: It is shown that the method gives nearly exact solutions.
38: %the absolute precision of the method is expected to be of order $10^{-5}$.
39:
40: %while the precision of the scaling approximation
41: % is about 0.5 \%.
42:
43: \end{quote}
44:
45: \vspace{5mm}
46: \noindent
47: PACS numbers: 03.75.-b, 03.75.Ss, 67.40.Db
48: \pagebreak
49:
50: \noindent
51: {\bf 1. Introduction}
52: \vspace{8pt}
53:
54: The newly created ultracold trapped Fermi gases with tunable atomic scattering
55: length [1-20] in the vicinity of a Feshbach resonance offer the possibility to
56: study highly
57: correlated many-body systems including the cross-over from the
58: Bardeen-Cooper-Schrieffer (BCS) phase to the Bose-Einstein condensate (BEC)
59: of molecules. Various investigations based on the hydrodynamic theory have
60: appeared recently [21-31].
61:
62: The purpose of this letter is to construct a simple, semi-analytical and
63: nearly exact formulae for hydrodynamic frequencies. Since
64: the collective frequencies can be measured with high precision,
65: these formulae will provide
66: a simple quantitative tool for the analysis of experimental data in
67: the hydrodynamic regime.
68:
69:
70: \vspace{8pt}
71:
72: \noindent
73: {\bf 2. Hydrodynamic theory}
74: \vspace{8pt}
75:
76: Our starting point is the quantum hydrodynamic theory [21-23]
77: for a dilute two-component Fermi gas in a
78: trap potential $V_{ext}(\vec{r})=(m/2)(\omega_{\perp}^2 (x^2+y^2)+
79: \omega_z^2 z^2)$
80: $$
81: i\hbar \frac{\partial \Psi}{\partial t}=-\frac{ \hbar^2}{2 m} \nabla^2 \Psi
82: +V_{ext} \Psi+V_{xc}\Psi,
83: \eqno{(1)}
84: $$
85: where
86: $$V_{xc}(\vec{r},t)=[\frac{\partial( n \epsilon(
87: n))}{\partial n}]_{n=n(\vec{r},t)},$$ $\epsilon$
88: is the ground state energy per particle of the homogeneous system and $n$ is the
89: density, $n(\vec{r},t)=\mid \Psi(\vec{r},t) \mid^2$,
90: normalized to the total number of atoms, $\int n(\vec{r},t) d^3r=N$.
91: It is useful to rewrite Eq.(1) in a form
92: $$
93: \frac{\partial n}{\partial t}+\nabla (n \vec{\mbox{v}})=0,
94: \eqno{(2)}
95: $$
96: $$
97: \frac{\partial \vec{\mbox{v}}}{\partial t}+\frac{1}{m}\nabla (V_{ext}+
98: \frac{d(n \epsilon(
99: n))}{d n}+\frac{1}{2} m \mbox{v}^2-\frac{\hbar^2}{2 m} \frac{1}{\sqrt{n}}
100: \nabla^2 \sqrt{n})=0,\eqno{(3)}
101: $$
102: where $\vec{\mbox{v}}$ is the velocity field, which for
103: $\Psi=e^{i \phi(\vec{r},t)} n^{1/2}(\vec{r},t)$ can be written as
104: $\vec{\mbox{v}}=(\hbar/m) \nabla \phi$.
105:
106: It can be proved [21] that every solution of the Eqs.(2,3) is a stationary point corresponding to the Lagrangian density
107: $$
108: \mathcal{L}_0=\hbar \dot{\phi}n+\frac{\hbar^2}{2m} (\nabla \sqrt{n})^2+
109: \frac{\hbar^2}{2m}n (\nabla \phi)^2+\epsilon(n)n+V_{ext}n.
110: \eqno{(4)}
111: $$
112: It was shown in Refs.[21-23] that for experimental conditions of
113: Refs.[14,16,17]
114: the quantum pressure term in Eqs.(3,4) can be neglected. For the reminder of
115: this
116: Letter we
117: will use this hydrodynamic approximation.
118: For the harmonic trap a trial function in the scaling ansatz is taken as
119: [21-23,26,28]
120: $$\phi(\vec{r},t)=\phi_0(t)+(m/(2 \hbar) \sum_{i=1}^3 \beta_i(t) x_i^2,$$
121: $$n(\vec{r},t)=n_0(x_i/b_i(t))/\kappa(t),$$ where $\kappa(t)=\prod_jb_j$ and
122: the Hamilton principle, $\delta\int dt\int\mathcal{L}_0 d^3 r=
123: 0$, gives the following equations for the scaling parameters $b_i$ [21-22]
124: $$
125: \ddot{b}_i+\omega_i^2(t) b_i-\frac{\omega_i^2}{b_i} \frac
126: {\int[n^2d\epsilon(n)/dn]_{n=n_0(\vec{r})/\kappa(t)} d^3 r}
127: {\int[n^2d\epsilon(n)/dn]_{n=n_0(\vec{r})} d^3 r}\kappa(t)=0,
128: \eqno{(5)}
129: $$
130:
131: Expanding Eqs.(5) around equilibrium ($b_i=1$) leads
132: to
133: the following result for the $M=0$ modes frequencies, $\omega^{(s)}$
134: in the scaling approximation
135: $$
136: \omega^{(s)}_{\pm}=\frac{\omega_{\perp}}{\sqrt{2}}
137: [\eta_s \pm \sqrt{\eta_s^2-8 \lambda^2 (3 \zeta_s+5)}]^{1/2},
138: \eqno{(6)}
139: $$
140: where $\eta_s=4+2\zeta_s+3 \lambda^2+\zeta_s \lambda^2$, $\zeta_s=
141: \int n_0^3d^2\epsilon/(dn_0^2)d^3r/\int n_0^2d\epsilon/(d n_0)d^3r$,
142: $\lambda=\omega_z/\omega_{\perp}$, and $\pm$ refer to the transverse and axial
143: mode, respectively.
144:
145: The hydrodynamic equations after linearization take the form
146: $$
147: \frac{\partial^2}{\partial t^2} \delta n+\frac{1}{m} \nabla (
148: n_0 \nabla (\frac{d^2(n_0 \epsilon(n_0)))}{d n_0^2} \delta n))=0,
149: \eqno{(7)}
150: $$
151: where $\delta n(\vec{r},t)$ is the change in the density profile with respect
152: to the equilibrium configuration. If we consider oscillations with time
153: dependence $\delta n \propto \exp(i \omega t)$, Eq.(7) can be reduced to
154: a Hermitian equation [27]
155: $$
156: \omega^2 [\frac{d^2(n_0 \epsilon(n_0))}{d n_0^2}]^{-1} |f>=L|f>
157: %\frac{1}{m} \nabla(n_0\nabla f),
158: \eqno{(8)}
159: $$
160: where $|f>=\frac{d^2(n_0 \epsilon(n_0))}{d n_0^2}|\delta n>$,
161: $L=-\frac{1}{m} \nabla n_0\nabla$
162: and the
163: equilibrium density, $n_0$, is given by equation
164: $$
165: \mu=V_{ext}+\frac{d(n_0 \epsilon(n_0))}{d n_0},
166: \eqno{(9)}
167: $$
168: where $\mu$ is the chemical potential, in the region where $n_0(\vec{r})$ is
169: positive and $n_0(\vec{r})=0$ outside this region.
170:
171: We note here that $d^2(n_0 \epsilon(n_0))/d n_0^2$ is positive, since the sound velocity for the homogeneous case is given by $c^2=(n_0/m)
172: d^2(n_0 \epsilon(n_0))/d n_0^2$.
173: \vspace{8pt}
174:
175: \noindent
176: {\bf 3. Equation of state}
177: \vspace{8pt}
178:
179:
180: For the negative S-wave scattering length between the two fermionic species,
181: $a<0$,
182: in the low-density regime, $k_F\mid a \mid \ll 1$, the ground state energy per
183: particle , $\epsilon(n)$, is well represented by an expansion in power of
184: $k_F \mid a \mid$ [32]
185: $$
186: \epsilon(n)=2 E_F[\frac{3}{10}-
187: \frac{1}{3 \pi} k_F \mid a \mid+0.055661 (k_F\mid a \mid)^2
188: -0.00914 (k_F\mid a \mid)^3+...],
189: \eqno{(10)}
190: $$
191: where $E_F=\hbar^2 k_F^2/(2 m)$ and $k_F=(3 \pi^2 n)^{1/3}$.
192: In the opposite regime, $a\rightarrow - \infty$
193: (the Bertsch many-body problem, quoted in
194: Refs.[33]), $\epsilon(n)$ is proportional to that of the non-interacting Fermi
195: gas
196: $$
197: \epsilon(n)=(1+\beta)\frac{3}{10} \frac{\hbar^2 k_F^2}{m},
198: \eqno{(11)}
199: $$
200: where a universal parameter $\beta$ [10] is estimated to be $\beta=-0.56$
201: [34]. The universal limit [10,34-37] is valid at least in the case where the
202: width of the Feshbach resonance is large compared to the Fermi energy as in
203: the cases of $^6Li$ and $^{40}K$.
204:
205: In the $a\rightarrow +0$ limit the system reduces to the
206: dilute Bose gas of dimers
207: $$\epsilon(n)=E_F(-1/(k_F a)^2+a_m k_F/(6 \pi)+...,
208: )
209: \eqno{(12)}
210: $$
211: where $a_m$ is the boson-boson scattering length, $a_m \approx 0.6 a$ [38].
212:
213: A simple interpolation of the form $\epsilon(n)\approx
214: E_F P(
215: k_F a)$ with a smooth function $P(x)$ was considered in several papers.
216: In Ref.[21] a [2/2] Pad\'{e} approximant
217: has been proposed for the function
218: $P(x)$ for the the case of negative $a$
219: $$
220: P(x)=\frac{3}{5}-2\frac{\delta_1\mid x \mid+\delta_2 x^2
221: }{1+\delta_3\mid x\mid+\delta_4 x^2},
222: \eqno{(13)}
223: $$
224: where $\delta_1=0.106103$, $\delta_2=0.187515$, $\delta_3=2.29188$,
225: $\delta_4=1.11616$.
226: Eq.(13) is constructed to reproduce the first four terms of the expansion (10)
227: in
228: the low-density regime and
229: also to exactly reproduce
230: results of
231: the recent Monte Carlo calculations [34], $\beta=-0.56$, in the unitary limit,
232: $k_F a \rightarrow -\infty$.
233:
234: For the positive $a$ case ( the interaction is strong enough to form bound
235: molecules
236: with energy $E_{mol}$) we have considered in Ref.[22] a [2/2] Pad\'{e}
237: approximant
238: $$
239: P(x)=\frac{E_{mol}}{2 E_F}+\frac{\alpha_1 x+\alpha_2 x^2}{1+\alpha_3 x+\alpha_4
240: x^2},
241: \eqno{(14)}
242: $$
243: where
244: parameters $\alpha_i$ are fixed by two continuity conditions at large $x$,
245: $1/x\rightarrow 0$, and by two continuity conditions at small $x$,
246: $\alpha_1=0.0316621$, $\alpha_2=0.0111816$, $\alpha_3=0.200149$,
247: and $\alpha_4=0.0423545$.
248:
249: In Ref.[39] a Pad\'{e} approximation has been considered for the
250: chemical potential. Authors of Ref.[28] have used a model for $P(x)$,
251: interpolating the Monte Carlo results of Ref.[36] across the unitary limit
252: and limiting behaviors for small $|x|$. We note here also the BCS mean-field
253: calculations of Ref.[28].
254: \vspace{8pt}
255:
256: \noindent
257: {\bf 4. Schwinger variational principle}
258: \vspace{8pt}
259:
260: The Schwinger variational principle (SVP) [40-43] can be generalized to the
261: case of
262: Eq.(8).
263:
264: Since the entire treatment is based on the equivalence of the SVP and the method of separable representation, we briefly describe this method. Let us consider the symbolic identity
265: $$
266: L=L L^{-1} L=\sum_{i,j} L|i><i|L^{-1}|j><j|L,
267: \eqno{(15)}
268: $$
269: where $|i>$ is a complete set. Truncating the summation over the complete set we obtain a separable approximation
270: $$
271: L^{(q)}=\sum_{i,j}^q L|\chi_i>
272: d_{ij}^{-1} <\chi_j|L,
273: \eqno{(16)}
274: $$
275: where $d_{ij}=<\chi_i|L|\chi_j>$.
276:
277: We note that Eq.(16) represents an interpolation process, since
278: $L^{(q)}|\chi_i>=L|\chi_i>$ and $<\chi_j|L^{(q)}=<\chi_j|L$.
279:
280: Substituting $L^{(q)}$ from Eq.(16) into Eq.(8), we obtain
281: $$
282: \omega^2 [\frac{d^2(n_0 \epsilon(n_0))}{d n_0^2}]^{-1} |f>=L^{(q)}|f>=
283: \sum_{i,j}^q L|\chi_i>
284: d_{ij}^{-1} <\chi_j|L|f>.
285: \eqno{(17)}
286: $$
287: We seek a solution of Eq.(17) in the form
288: $$
289: |f>=\sum_i^q c_i \frac{d^2(n_0 \epsilon(n_0))}{d n_0^2} L|\chi_i>,
290: \eqno{(18)}
291: $$
292: then $c_i$'s are defined from equations
293: $$
294: \sum_{k=1}^q B_{ik}(\omega^2) c_k=0,
295: \eqno{(19)}
296: $$
297: where
298: $$
299: B_{ik}(\omega^2)=<\chi_i|(\omega^2 L-L\frac{d^2(n_0 \epsilon(n_0))}{d n_0^2} L)
300: |\chi_k>
301: \eqno{(20)}
302: $$
303: and frequencies $\omega$ are determined from the condition of vanishing of the determinant of the matrix $B_{ik}(\omega^2)$.
304: $$
305: det B_{ik}(\omega^2)=0
306: \eqno{(21)}
307: $$
308:
309: Defining the Schwinger functional $I_{SVP}[\chi] $ by
310: $$
311: I_{SVP}[\chi]= \frac{<\chi|L \frac{d^2(n_0 \epsilon(n_0))}{d n_0^2} L|\chi>}{<\chi| L|\chi>},
312: \eqno{(22)}
313: $$
314: we get $I_{SVP}[f]=\omega^2$, where $f$ is the
315: solution of Eq.(8). Introducing the function $\chi=\sum_{r=1}^q c_r \chi_r$,
316: where $c_1,c_2, ... c_q
317: $ are $q$ variable parameters, we see that the functional $I_{SVP}$ is
318: stationary if
319: $\partial I_{SVP}/\partial c_r=0$. The later equations coincide with Eqs.(19),
320: and therefore the approximate solution of the Eq.(8), which is based on the SVP, is equivalent to the exact solution of the Eq.(8) with separable $L^{(q)}$.
321:
322: Operator $L$ is clearly positive, which means that $<u|L|u>\geq0$ for all $u$,
323: but is not positive definite, since $<u|L|u>=0$ for some $u\neq 0$.
324: It can easily be seen that
325: $$<u|(L-L^{(q)})|u>\geq 0,
326: \eqno{(23)}
327: $$
328: for all $u$.
329: Indeed,
330: $$
331: J=<(u+\sum_{i=1}^qc_i \chi_i)|L|(u+\sum_{j=1}^q c_j \chi_j)> \geq 0
332: $$
333: for all $c_i$. We choose the $c_i$'s from the conditions
334: $\partial J/\partial c_i=0$, and then obtain
335: $$
336: J=<u|(L-L^{(q)})|u>\geq0
337: $$
338: for all $q$.
339:
340: Since the problem is solved by replacing $L$ by $L^{(q)}$ and since the
341: operator $(L-L^{(q)})$ is positive, the SVP leads to the approximate lower
342: bounds for $\omega$ up to the second order of $|L-L^{(q)}|$.
343:
344: For the most interested case of $M=0$ modes, we can put in Eq.(16)
345: $q=2$,$~$
346: $~$ $\chi_1=(x^2+y^2)$ and $\chi_2=z^2$, which give
347: $$
348: \omega^{SVP}_{\pm}=\frac{\omega_{\perp}}{\sqrt{2}}
349: [\eta_{SVP} \pm \sqrt{\eta_{SVP}^2- 8 \lambda^2 (9 \zeta_{SVP}-1)}]^{1/2},
350: \eqno{(24)}
351: $$
352: where $\eta_{SVP}= 6 \zeta_{SVP}+ \lambda^2 (3 \zeta_{SVP}+1)$,
353: $\zeta_{SVP}=-\int n_0^2 \tilde{x}^3 [\frac{\partial n_0}{\partial \tilde{x}}]^{-1}
354: d \tilde{x}/\int n_0 \tilde{x}^4 d \tilde{x}$,
355: $\tilde{x}=\sqrt{x^2+y^2+\lambda^2 z^2}$ and $\pm$ signs refer to the
356: transverse and axial mode, respectively.
357:
358: It is easy to show that Eq. (24) gives exact
359: solutions for frequencies of the breathing modes for the polytropic equation
360: of state, $\epsilon(n)\approx n^{\gamma}$.
361: In Ref.[24], on the basis of a generalization of the
362: Hylleraas-Undheim method, we have constructed rigorous upper bounds to the
363: collective
364: frequencies for the radial and the axial breathing mode of the Fermi gas
365: under harmonic confinement in the framework of the
366: hydrodynamic theory
367: $$
368: \omega^{upper}_{\pm}=\frac{\omega_{\perp}}{\sqrt{5 \zeta_{upper}-9}}
369: [\eta_{upper} \pm \sqrt{
370: \eta_{upper}^2-8 \lambda^2 \zeta_{upper} (5 \zeta_{upper}-9)}]^{1/2},
371: \eqno{(25)}
372: $$
373: where $\eta_{upper}= (3+4 \lambda^2) \zeta_{upper} - (3+6 \lambda^2)$,
374: $\zeta_{upper}=I_0 I_4/I_2^2,$ and
375: $I_l=\int \tilde{x}^l n_0(\tilde{x}) d\tilde{x}$.
376:
377:
378: We expect that the difference between $\omega^{SVP}_{\pm}$ and
379: $\omega^{upper}_{\pm}$ characterizes the error and is not very sensitive to
380: the functional form of $\epsilon(n)$. From Table 1 one can see
381: that this difference is order of $10^{-5}$. We note that the comparison with
382: the
383: scaling approximation, Eq.(6), shows that the absolute precision of the
384: scaling approximation is about $10^{-3}$ that agrees with Ref.[31].
385:
386: In Fig.1, we have compared the hydrodynamic predictions for $\omega_+$
387: with experimental
388: data [17]. There is a very good agreement with experimental data [17] near
389: the unitary limit. We note here that two
390: experimental results [17] and [16] (not shown in Fig.1)
391: for $\omega_+$ are still
392: about 10\% in disagreement with
393: each other, which is not fully understood yet.
394:
395: To calculate $\zeta$'s, we have used the very fast converged expansion of
396: Ref.[24]
397: $$
398: n_0(\vec{r})\approx (1-\beta V_{ext}(\vec{r}))^{1/(2-p)}
399: \sum_{i=0}^{l-1}\tilde{c}_{i} [V_{ext}(\vec{r})]^i,
400: $$
401: where parameters $\beta$, $p$ and $\tilde{c}_i$ are fixed by requiring that
402: $n_0(\vec{r})$ must satisfy a variational principle
403: $\delta \int n_0 (V_{ext}+\epsilon(n_0))d^3 r=0$ with a subsidiary condition
404: $\int n_0 d^3 r=N$.
405: \vspace{8pt}
406:
407: \noindent
408: {\bf 4. Summary}
409: \vspace{8pt}
410:
411: We have generalized the Schwinger variational method for the trapped strongly interacting atoms in hydrodynamic regime and we have constructed semi-analytical
412: and extremely accurate formulae for hydrodynamic collective frequencies.
413: These formulae are very useful since they provide an easy and simple quantitive tool for the analysis of experimental data for trapped condensed gases without relying on complex and extensive computations.
414:
415: \pagebreak
416:
417: Table 1. The transverse and axial frequencies
418: in units of $\omega_{\perp}$ and $\omega_z$, respectively,
419: in the BCS region as a function
420: of the dimensional parameter
421: $X=(N^{1/6} a/a_{ho})^{-1}$. The trap parameter $\lambda$ is assumed to be
422: 0.045613.
423: The [2/2] Pad\'{e}
424: approximation
425: of Refs.[21,22] is used for the energy per particle $\epsilon(n)$.\\
426:
427: \begin{tabular}{lllllll}
428: \hline\hline
429: X
430: &$\omega^{upper}_+$
431: &$\omega^{SVP}_+$
432: &$\omega^{upper}_-$
433: &$\omega^{SVP}_-$
434: &$\omega^{s}_+$
435: &$\omega^{s}_-$ \\ \hline
436: -0.1
437: &1.8160
438: &1.8160
439: &1.5470
440: &1.5470
441: &1.8193
442: &1.5477 \\ \hline
443: -0.3
444: &1.8015
445: &1.8015
446: &1.5438
447: &1.5438
448: &1.8082
449: &1.5453 \\ \hline
450: -0.5
451: &1.7931
452: &1.7931
453: &1.5419
454: &1.5419
455: &1.8002
456: &1.5435 \\ \hline
457: -0.7
458: &1.7886
459: &1.7886
460: &1.5409
461: &1.5409
462: &1.7947
463: &1.5423 \\ \hline
464: -0.9
465: &1.7867
466: &1.7867
467: &1.5405
468: &1.5405
469: &1.7910
470: &1.5414 \\ \hline
471: -1.1
472: &1.7861
473: &1.7861
474: &1.5403
475: &1.5403
476: &1.7886
477: &1.5409 \\ \hline
478: -1.3
479: &1.7865
480: &1.7865
481: &1.5404
482: &1.5404
483: &1.7871
484: &1.5406 \\ \hline
485: -1.5
486: &1.7873
487: &1.7873
488: &1.5406
489: &1.5406
490: &1.7863
491: &1.5404 \\ \hline
492: -1.7
493: &1.7884
494: &1.7884
495: &1.5409
496: &1.5409
497: &1.7860
498: &1.5403 \\ \hline
499: -2.0
500: &1.7902
501: &1.7902
502: &1.5413
503: &1.5413
504: &1.7861
505: &1.5403 \\ \hline \hline
506: \end{tabular}
507:
508: \pagebreak
509: \begin{figure}[ht]
510: \includegraphics{svp.eps}
511: \end{figure}
512: Fig. 1.
513: Radial breathing mode frequency $\omega_+$ in the BCS region as a
514: function
515: of the dimensional parameter $X=(N^{1/6} a/a_{ho})^{-1}$ (the solid line).
516: The dashed line represent the scaling approximation, Eq.(6).
517: The circular dots
518: with error
519: bars are the experimental results given by the Duke University group [17].
520: Everything is measured in units of $\omega_{\perp}$.
521:
522: \pagebreak
523:
524: {\bf References}
525: \vspace{8pt}
526:
527: \noindent
528: [1] Gehm M E and Thomas J E 2004
529: {\it Am. Sci.} {\bf 92} 238
530:
531: \noindent
532: [2] Regal C A, Greiner M and Jin D S 2004 {\it Phys. Rev. Lett.}
533: {\bf 92} 040403
534:
535: \noindent
536: [3] Greiner M, Regal C A and Jin D S 2003 {\it Nature} {\bf 426} 537
537:
538: \noindent
539: [4] Stecker K E, Patridge G B and Hulet R G 2003
540: {\it Phys. Rev. Lett.} {\bf 91}
541: 080406
542:
543: \noindent
544: [5] Cubizolles J, Bourdel T, Kokkelmans S J J M F
545: Shlyapnikov G V and
546: Salomon C 2003 {\it Phys. Rev. Lett.} {\bf 91} 240401
547:
548: \noindent
549: [6] Jochim S, Bartenstein M, Altmeyer A, Hendl G, Chin C,
550: Denschlag J H ,
551: and
552: Grimm R 2003 {\it Phys. Rev. Lett.} {\bf 91} 240402
553:
554: \noindent
555: [7] Jochim S, Bartenstein M, Altmeyer A, Hendl G, Riedl S,
556: Chin C,
557: Denschlag J H and Grimm R 2003 {\it Science} {\bf 302} 2101
558:
559: \noindent
560: [8] Zwierlein M W, Stan C A, Schunck C H,
561: Raupach S M F, Gupta S,
562: Hadzibabic Z, Ketterle W 2003 {\it Phys. Rev. Lett.} {\bf 91} 250401
563:
564: \noindent
565: [9] Regal C A and Jin D S 2003 {\it Phys. Rev. Lett.} {\bf90}
566: 230404
567:
568: \noindent
569: [10] O'Hara K M, Hemmer S L, Gehm M E, Granade S R , and
570: Thomas J E 2002
571: {\it Science} {\bf 298} 2179
572:
573: \noindent
574: [11] Kinast J, Turlapov A, Thomas J E, Chen Q, Stajic J, and
575: Levin K 2005 {\it Science} {\bf 307} 1296
576:
577: \noindent
578: [12] Kinast J, Turlapov A and Thomas J E 2005 {\it Preprint}
579: cond-mat/0502507
580:
581: 2005 {\it Phys. Rev. Lett.} {\bf94} 170404
582:
583: \noindent
584: [13] Bartenstein M, Altmeyer A, Riedl S, Jochim S, Chin C,
585: Denschlag J H and Grimm R 2004 {\it Phys. Rev. Lett.} {\bf92} 120401
586:
587: \noindent
588: [14] Kinast J, Hemmer S L, Gehm M E, Turlapov A and Thomas J E 2004
589: {\it Phys. Rev. Lett.} {\bf 92} 150402
590:
591: \noindent
592: [15] Bourdel T, Khaykovich L, Cubizolles J, Zhang J, Chevy F,
593: Teichmann M, Tarruell L, Kokkelmans S J J M F and Salomon C 2004
594: {\it Phys. Rev. Lett.} {\bf 93} 050401
595:
596: \noindent
597: [16] Bartenstein M, Altmeyer A, Riedl S, Jochim S, Chin C,
598: Denschlag J H, and Grimm R 2004 {\it Phys. Rev. Lett.}
599: {\bf 92} 203201
600:
601: 2005 {\it Preprint} 2005 {\it Preprint} cond-mat/0412712
602:
603: \noindent
604: [17] Kinast J, Turlapov A, and Thomas J E 2004 {\it Phys. Rev.}
605: A {\bf 70}
606: 051401(R)
607:
608:
609: \noindent
610: [18] C. Chin C, Bartenstein M, Altmeyer A, Riedl S, Jochim S,
611: Denschlag J H and Grimm R 2004 {\it Science} {\bf 305} 1128
612:
613: \noindent
614: [19]
615: Zwierlein M W, Stan C A , Schunck C H, Raupach S M F, Kerman A J,
616: and Ketterle W 2004 {\ Phys. Rev. Lett.} {\bf 92} 120403
617:
618: \noindent
619: [20] Kinast J, Turlapov A, and Thomas J E, 2005 {\it Preprint}
620: cond-mat/0503620
621:
622: \noindent
623: [21] Kim Y E and Zubarev A L 2004 {\it Phys. Lett.} A{\bf 327} 397
624:
625: \noindent
626: [22] Kim Y E and Zubarev A L 2004
627: {\it Phys. Rev.} A
628: {\bf 70} 033612
629:
630: \noindent
631: [23] Kim Y E and Zubarev A L 2004 {\it Preprint} cond-mat/0408283
632:
633: \noindent
634: [24] Kim Y E and Zubarev A L 2005 {\it Preprint} cond-mat/0502651.
635:
636: \noindent
637: [25] Heiselberg H, 2004 {\it Phys. Rev. Lett.} {\bf 93} 040402
638:
639: \noindent
640: [26] Menotti C, Pedri P and Stringari S 2002 {\it Phys. Rev. Lett.}
641: {\bf 89} 250402
642:
643:
644: \noindent
645: [27] Bulgak A and Bertsch G F 2005 {\it Phys. Rev. Lett.}
646: {\bf 94} 070401
647:
648: \noindent
649: [28] Hu H, Minguzzi A, Liu Xia-Ji and Tosi M P 2004
650: {\it Phys. Rev. Lett.} {\bf 93} 190403
651:
652: \noindent
653: [29] Manini N and Salasnich L, 2005 {\it Phys. Rev.} A
654: {\bf 71} 033625
655:
656: \noindent
657: [30] Combescot R and Leyronas X, 2004 {\it Phys. Rev. Lett.}
658: {\bf 93} 138901
659:
660: 2004 {\it Europhys. Lett.} {\bf 68} 762
661:
662: \noindent
663: [31] Astrakharchik G E, Combescot R, Leyronas X and Stringari S
664: 2005 {\it Preprint}
665: cond-mat/0503618
666:
667: \noindent
668: [32] Lenz W 1929 {\it Z. Phys.} {\bf 56} 778
669:
670: Huang K and Yang C N 1957
671: {\it Phys. Rev.} {\bf105} 767
672:
673: Lee T D and Yang C N 1957 {\it ibid} {\bf 105} 1119
674:
675: Efimov V N and Amus'ya M Ya 1965
676: {\it Sov. Phys. JETP} {\bf 20}, 388
677:
678: \noindent
679: [33] Baker Jr. G A , 2001 {\it Int. J. Mod. Phys.} B{\bf15} 1314
680:
681: 1999 {\it Phys. Rev. C} {\bf
682: 60}
683: 054311
684:
685: \noindent
686: [34] Carlson J, Chang S Y, Pandharipande V R and
687: Schmidt K E 2003
688: {\it Phys. Rev. Lett.} {\bf
689: 91}
690: 050401
691:
692: \noindent
693: [35] Combescot R 2003 {\it Phys. Rev. Lett.} {\bf 91} 120401
694:
695: \noindent
696: [36] Astrakharchik G E, Boronat J, Casulleras J and
697: Giorgini S 2004
698: {\it Phys. Rev. Lett.} {\bf 93}, 200404
699:
700: \noindent
701: [37] Ho T L and Mueller E J 2004 {\it Phys. Rev. Lett.}
702: {\bf 92} 160404
703:
704: Ho T L 2004 {\it ibid} {\bf 92} 090402
705:
706: \noindent
707: [38] Petrov D S, Salomon C, and Shlyapnikov G V, 2004
708: {\it Phys. Rev. Lett.} {\bf 93} 090404
709:
710: \noindent
711: [39] Combescot R and Leyronas X, 2004 {\it Preprint} cond-mat/0407388
712:
713: \noindent
714: [40] Zubarev A L 1981 {\it The Schwinger Variational Principle in Quantum
715: Mechanics} (Moscow: Energoatomizdat) [in Russian]
716:
717: \noindent
718: [41] Zubarev A L 1976 {\it Sov. J. Part. Nucl.} {\bf 7} 215
719:
720: 1978 {\it ibid} {\bf 9}, 188
721:
722: \noindent
723: [42] Belyaev V B, Podkopayev A P, Wrzecionko J and
724: Zubarev A L 1979
725: {\it J. Phys.} B {\bf 12}, 1225
726:
727: \noindent
728: [43] Watson D K 1988 {\it Adv. in Atom. Mol. Phys.} {\bf 25} 221
729:
730: \end{document}
731: