cond-mat0506264/Dunk.tex
1: 
2: %\usepackage{amsfonts,latexsym}
3: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
4: %    INSTITUTE OF PHYSICS PUBLISHING                                   %
5: %                                                                      %
6: %   `Preparing an article for publication in an Institute of Physics   %
7: %    Publishing journal using LaTeX'                                   %
8: %                                                                      %
9: %    LaTeX source code `ioplau2e.tex' used to generate `author         %
10: %    guidelines', the documentation explaining and demonstrating use   %
11: %    of the Institute of Physics Publishing LaTeX preprint files       %
12: %    `iopart.cls, iopart12.clo and iopart10.clo'.                      %
13: %                                                                      %
14: %    `ioplau2e.tex' itself uses LaTeX with `iopart.cls'                %
15: %                                                                      %
16: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
17: %
18: %
19: % First we have a character check
20: %
21: % ! exclamation mark    " double quote
22: % # hash                ` opening quote (grave)
23: % & ampersand           ' closing quote (acute)
24: % $ dollar              % percent
25: % ( open parenthesis    ) close paren.
26: % - hyphen              = equals sign
27: % | vertical bar        ~ tilde
28: % @ at sign             _ underscore
29: % { open curly brace    } close curly
30: % [ open square         ] close square bracket
31: % + plus sign           ; semi-colon
32: % * asterisk            : colon
33: % < open angle bracket  > close angle
34: % , comma               . full stop
35: % ? question mark       / forward slash
36: % \ backslash           ^ circumflex
37: %
38: % ABCDEFGHIJKLMNOPQRSTUVWXYZ
39: % abcdefghijklmnopqrstuvwxyz
40: % 1234567890
41: %
42: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
43: 
44: \documentclass[12pt]{iopart}
45: % Uncomment next line if AMS fonts required
46: %\usepackage{iopams}
47: \usepackage{graphicx,latexsym,stmaryrd,iopams}
48: 
49: % remove me after proof readin - color only for correction \#\#\#
50: \usepackage{color}
51: 
52: \begin{document}
53: 
54: \title{Exact results for the $1$D interacting Fermi gas with arbitrary polarization}
55: 
56: \author{M.T. Batchelor$^{1,2}$, M. Bortz$^1$, X.W. Guan$^{1,2}$ and N. Oelkers$^{1,2}$}
57: \address{$^1$
58: {\small Department of Theoretical Physics, Research School of Physical Sciences and Engineering}}
59: \address{$^2$
60: {\small Department of Mathematics, Mathematical Sciences Institute\\}
61: {\small The Australian National University, Canberra ACT 0200, Australia}}
62: 
63: %\maketitle
64: 
65: \begin{abstract}
66: %
67: We investigate the $1$D interacting two-component Fermi gas with arbitrary polarization. 
68: %
69: Exact results for the ground state energy, quasimomentum distribution functions, spin velocity and
70: charge velocity reveal subtle polarization dependent quantum effects.  
71: %
72: \end{abstract}
73: 
74: \pacs{03.75.Ss, 05.30.Fk, 71.10.Pm}
75: 
76: \maketitle
77: 
78: 
79: Recent achievements in trapping quantum gases of ultra-cold atoms have
80: opened up many exciting possibilities for the experimental
81: investigation of quantum effects in low-dimensional many-body systems.
82: %
83: Among these, progress in realizing interacting Fermi gases 
84: %via Feshbach resonance 
85: promises new opportunities for studying the pronounced crossover from a
86: Bardeen-Cooper-Schrieffer (BCS) superfluid to a Bose-Einstein
87: condensate (BEC) \cite{BEC-F}.
88: %
89: A crossover from weakly attractive Cooper pairs  to strongly repulsive 
90: molecular dimers can also take place in a confined quasi $1$D Fermi gas of atoms  
91: by using magnetic field induced Feshbach resonances \cite{Fermi-1D1,Fermi-1D2}. 
92: %
93: In this scenario, a number of key $1$D integrable models with a
94: variable interaction parameter have been used to describe the crossover 
95: \cite{Fermi-1D2,BEC-BCS2,BEC-BCS3,BEC-BCS1,BEC-BCS4,BEC-BCS5,BEC-BCS6}.
96: %
97: These include the integrable interacting Fermi gas \cite{Gaudin-Yang}. 
98: 
99: 
100: The $1$D Fermi gas can be experimentally realized \cite{Fermi-1D1} by tightly confining 
101: the atomic cloud in the radial directions and weakly confining it along the axial direction
102: \cite{Fermi-1D2,BEC-BCS2,BEC-BCS3,BEC-BCS1,SC-S}. 
103: %
104: To date, the theoretical study of the $1$D continuum Fermi gas has been mainly 
105: for the zero polarization case in the thermodynamic limit (TL).
106: %
107: Our aim here is to present the results of a comprehensive study of the $1$D
108: exactly solved two-component Fermi gas with arbitrary polarization, throughout
109: the whole interaction range. 
110: %
111: This includes finite systems in the weak and strong coupling regions, and in the TL.
112: 
113: {\it The model.} 
114: The Hamiltonian \cite{Gaudin-Yang}   
115: %
116: \begin{equation}
117: {H}=-\frac{\hbar ^2}{2m}\sum_{i = 1}^{N}\frac{\partial
118: ^2}{\partial x_i^2}+\,g_{\rm 1D} \sum_{1\leq i<j\leq N} \delta
119: (x_i-x_j)
120: \label{Ham-1}
121: \end{equation}
122: %
123: describes a $\delta$-function interacting gas of $N$ fermions each of mass $m$
124: constrained by periodic boundary conditions to a line of length $L$.
125: %
126: The coupling constant $g_{\rm 1D}$ can be written in terms of 
127: the scattering strength $c={2}/{a_{\rm 1D}}$ as $g_{\rm 1D} ={\hbar ^2 c}/{m}$.
128: %
129: An effective $1$D scattering length $a_{\rm 1D}$ can be expressed through the 
130: $3$D scattering length for fermions confined in a $1$D geometry. 
131: %
132: We use the dimensionless coupling constant $\gamma={mg_{\rm 1D}}/{(\hbar^2n})$ 
133: for physical analysis. 
134: %
135: Here $n={N}/{L}$ is the linear density. 
136: %
137: The coupling $g_{\rm 1D}$ can vary from $-\infty$ to $\infty$ by 
138: detuning the $3$D scattering length.
139: %
140: The interaction is attractive for $g_{\rm 1D}<0$ and repulsive for $g_{\rm 1D}>0$.  
141: 
142: 
143: The energy is given by $E=\frac{\hbar ^2}{2m}\sum_{j=1}^Nk_j^2$, where 
144: the quasimomenta $k_j$ satisfy the Bethe equations (BE) \cite{Gaudin-Yang}
145: %
146: %Namely, $\exp({\rm i}k_jL)=f_\beta^M (k_j, \Lambda_\beta, \frac12 c)$ and
147: %$f_\ell^N (\Lambda_\alpha, k_\ell, \frac12 c) = - f_\beta^M (\Lambda_\alpha, \Lambda_\beta, c)$,
148: %for $j = 1,\ldots, N$ and $\alpha = 1,\ldots, M$, with $M$ the number of spin-down fermions.
149: %
150: %Here $f_n^m(p,q,r)=\prod^m_{n = 1} \frac{p-q+\mathrm{i}r}{p-q-\mathrm{i}r}$.
151: %
152: \begin{eqnarray}
153: & &\exp(\mathrm{i}k_jL)=\prod^M_{\ell = 1} 
154: \frac{k_j-\Lambda_\ell+\mathrm{i}\, c/2}{k_j-\Lambda_\ell-\mathrm{i}\, c/2},\nonumber\\
155: & &\prod^N_{\ell = 1}\frac{\Lambda_{\alpha}-k_{\ell}+\mathrm{i}\, c/2}{\Lambda_{\alpha}-k_{\ell}-\mathrm{i}\, c/2}
156:  = - {\prod^M_{ \beta = 1} }
157: \frac{\Lambda_{\alpha}-\Lambda_{\beta} +\mathrm{i}\, c}{\Lambda_{\alpha}-\Lambda_{\beta} -\mathrm{i}\, c} .
158: \label{BE}
159: \end{eqnarray}
160: %
161: Here $j = 1,\ldots, N$ and $\alpha = 1,\ldots, M$, with $M$ the number of spin-down fermions.
162: %
163: For convenience we take $N$ even.
164: %
165: Of particular interest are the bound states for attractive interaction. 
166: %
167: In this regime, the model (\ref{Ham-1}) has been extensively studied with $N=2M$ 
168: in the context of the BCS-BEC crossover \cite{Fermi-1D2,BEC-BCS2,BEC-BCS3,BEC-BCS1,BEC-BCS4}.  
169: %
170: For weak attraction, the system describes weakly bound Cooper pairs. 
171: %
172: In the strongly attractive regime the bound states behave like bosonic 
173: tightly bound molecular dimers. 
174: %
175: On the other hand, for repulsive interaction, there are no bound states for the ground state. 
176: %
177: The spin-up and spin-down fermions behave like indistinguishable free fermions
178: as $c\to \infty$. 
179: %
180: The ground state energy per particle in the TL,  i.e., $N, M, L \to \infty$, $N/L, M/L$ constant, 
181: can be derived from the Gaudin integral equations \cite{Gaudin-Yang}. 
182: %
183: Here we first analyze the BE directly, i.e., for a finite system. 
184: 
185: 
186: {\it The ground state.}
187: % 
188: We observe that for weak attractive interaction a pair of fermions with spin-up and 
189: spin-down states forms a bound state. 
190: %
191: On the other hand, for weak repulsive interaction the fermion momenta 
192: separate  in quasimomentum space. 
193: %
194: For $L|c| \ll 1$, it is convenient to distinguish between unpaired 
195: $k_j^{(\rm u)}$ ($j=1,\ldots,N-2M$) and paired $k_j^{(\rm p)}$ ($j=1,\ldots,2M$) momenta. 
196: %
197: On expanding the BE in powers including $\mathcal{O}(c)$ \cite{BGM}, 
198: one obtains $k_j^{(\rm u)}=\pi(M-N-1+2j)/L+\delta_j^{(\rm u)},\,j=1,\ldots,N/2-M$, $k_j^{(\rm u)}=\pi(3M-N-1+2j)/L+\delta_j^{(\rm u)},\,j=N/2-M+1,\ldots,N-2M$,  and 
199: $k_j^{(\rm p)}=\pi \left(-1-M+2j_+\right)/L+ \delta^{(\rm  p)}_{j_+}\pm \sqrt{c/L}$, $(j=1,\ldots,M)$,
200: where $j_+=j$, if $j$ odd and $j_+=j-1$ if $j$ even. Therefrom it follows that the binding energy in the weakly attractive case is $c/L$ for finite systems.
201: %
202: The deviations $\delta$ from $k_j^{(\rm u,p)}$ are linear in $c$, with
203: %
204: \begin{eqnarray}
205: \delta_j^{(\rm u)}&=&\frac cL {\sum_\ell}' \frac{1}{k_{j,0}^{(\rm u)}
206: -k_{\ell,0}^{(\rm p)}},\label{del1}\\
207: \delta_j^{(\rm p)}&=&\frac cL \left[{\sum_\ell}' \frac{1}{k_{j,0}^{(\rm p)}
208: -k_{\ell,0}^{(\rm p)}}+\frac{1}{2}\sum_{\ell}\frac{1}{k_{j,0}^{(\rm p)}
209: -k_{\ell,0}^{(\rm u)}}\right]\label{del2},
210: \end{eqnarray}
211: %
212: where the primed sum counts each $k_{\ell,0}^{(\rm p)}$ only once and excludes $k_{j,0}^{(\rm p)}$.
213: %
214: Here $k_{j,0}^{(\rm u,p)}$ are the quasimomenta of non-interacting particles 
215: as given above.  
216: %
217: Note that the unpaired momenta do not interact with each other, which is consistent with the Pauli principle.  
218: %
219: On the other hand, for strong attractive interaction, i.e., $Lc\ll -1$,
220: tightly bound states with quasimomentum 
221: $k_i^{(\rm p)} \approx \Lambda_i \pm \frac{1}{2} {\mathrm i} c$  
222: separate from unbound states with quasimomentum
223: $k_j^{(\rm u)}=\frac{n_j\pi}{L}\left(1+\frac{4M}{Lc}\right)^{-1}$ with integers 
224: $n_j=\pm 1, \pm 3,\ldots, \pm(N-2M-1)$. 
225: %
226: In the above, $\Lambda_i=\frac{n_i\pi}{L}\left(1+\frac{M}{Lc}+\frac{2(N-2M)}{Lc}\right)^{-1}$
227: for $n_i=\pm 1,\ldots, \pm \frac12{(M-1)}$. 
228: %
229: In this scenario the bound states behave like
230: hard-core bosons due to Fermi pressure.
231: 
232: 
233: 
234: 
235: We now define the fraction $a={2M}/{N}$ which characterizes the polarization of the 
236: system.\footnote{The polarization dependence of atom-hole condensation in optical lattices was discussed
237: in Ref.~\cite{Lee}.}
238: %
239: The ground state energy per particle in the weakly interacting regime follows from the 
240: above asymptotic roots of the BE as (in units of $\hbar^2n^2/(2m)$, with 
241: $a_0=(1-a)\left((1-\frac{1}{2}a)^2+\frac{1}{2}a\right)$)
242: %
243: \begin{eqnarray}
244: \frac{E}{N}&=& 
245: \frac{a^2}{2}\gamma+\gamma(1-a)a+\frac{\pi^2}{12}a^3+\frac{\pi^2}{3}a_0
246: +\mathcal{O}(\gamma^2).\label{E-W}
247: \end{eqnarray}
248: %
249: The first two terms represent the binding and the interaction energy, respectively, which are both linear in $\gamma$. 
250: %
251: This is different from what is obtained in the TL from the Gaudin
252: integral equations \cite{BEC-BCS2,BEC-BCS1,BEC-BCS4}, where the binding energy is $\propto \gamma^2$. 
253: %
254: The other terms in (\ref{E-W}) are the kinetic energies for paired and unpaired
255: fermions. 
256: %
257: In the special case $a=0$, the model reduces to fully polarized free fermions without $s$-wave scattering. 
258: %
259: Using the given asymptotic expressions for the Bethe roots in the strongly 
260: attractive regime, we obtain (in units of $\hbar^2n^2/(2m)$)
261: %
262: \begin{eqnarray}
263: \frac{E}{N}&=& 
264: -\frac{a}{4}\gamma^2+ \frac{\pi^2}{3} \left( \frac{a^3}{16} (1 - 2b_1) +
265: a_1(1-2b_2) \right) + \mathcal{O}(1/\gamma^2) , \label{E-S-A}
266: \end{eqnarray}
267: %
268: where $a_1=(1-a)^3$, 
269: $b_1 = a/(2\gamma) + 2(1-a)/\gamma$ and
270: $b_2 = 2a/\gamma$.
271: %
272: Whereas the first term represents the binding energy, the remaining interaction energies stem from
273: the interaction between pair-pair and pair-unpaired fermions.  
274: %
275: The ground state energy is lowest for $a=1$, compared to the others for $0\le a<1$. 
276: %
277: This is in general agreement with the Lieb-Mattis theorem \cite{LM-thorem}.
278: 
279: 
280: 
281: Now consider the crossover to the TL. 
282: %
283: First we observe that in the TL, the results \eref{E-W}, \eref{E-S-A} %
284: reduce to those derived
285: from the Gaudin integral equations \cite{BEC-BCS2,BEC-BCS3,BEC-BCS1,BEC-BCS4}.  
286: %
287: Furthermore, from \eref{del1}, \eref{del2}, we obtain the quasimomentum distribution functions 
288: $n_{\rm u,p}(k)$ per unit length in the TL, 
289: %
290: \begin{eqnarray}
291: \!\!\!n_u(k)&=& \frac{1}{2\pi} +\frac{c}{2\pi^2}\,\frac{A}{k^2-A^2},\; A<|k|<B,\label{n1}\\
292: \!\!\!n_p(k)&=& \frac1\pi -\frac{c}{2\pi^2}\left(\frac{A}{A^2-k^2}+\frac{B}{B^2-k^2}\right),\, |k|<A,\label{n2}
293: \end{eqnarray}
294: with $A=\pi M/L$ and $B=\pi (N+M)/L$. 
295: %
296: Especially, for $M=N/2$, $n_{\rm u}\equiv 0$.
297: %
298: In Fig. \ref{Fig1}, $n_{u,p}(k)$ as calculated from the finite BE are compared to Eqs. \eref{n1}, \eref{n2}. 
299: 
300: 
301: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
302: \begin{figure}[t]
303: \begin{center}
304: \includegraphics[width=.80\linewidth]{repatt.eps}
305: \end{center}
306: \caption{
307: The quasimomentum distribution function (crosses) for a finite system ($N=50, M=16$) 
308: for the a) weakly repulsive and b) weakly attractive regime. 
309: %
310: The full lines connecting the crosses are guides to the eye. 
311: %
312: The insets are zooms of the distribution near the noninteracting cases 
313: (dotted lines), $1/2\pi$ for unpaired fermions and $1/\pi$ for paired fermions.
314: %
315: Here the full lines are the results obtained in the TL (Eqs.~\eref{n1}, \eref{n2}).
316: }
317: \label{Fig1}
318: \end{figure}
319: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
320: 
321: 
322: We now use the Gaudin integral equations to numerically compute the dependence of 
323: $E/N$ on $\gamma$ for different $a$. 
324: %
325: The result is shown in Fig. \ref{Fig2}. 
326: %
327: The numerical data agree with \eref{E-W} and \eref{E-S-A} in the weakly 
328: interacting and strongly attracting regimes. 
329: %
330: A straightforward analysis of the integral equations in the strongly repulsive regime leads to 
331: %
332: \begin{eqnarray}
333: \frac{E}{N}\approx \left\{\begin{array}{ll}
334: \frac{\hbar^2n^2}{2m}\frac{\pi^3}{3}\left(1 - \frac{4a}{\gamma}\right)
335: +\mathcal{O}(1/\gamma^2,a^2/\gamma),&\,a\ll1\\
336:  \frac{\hbar^2n^2}{2m}\frac{\pi^3}{3}\left(1-\frac{4 \ln 2}{\gamma}\right)
337:  +\mathcal{O}(1/\gamma^2),&\,a=1
338: \end{array}
339: \right.\label{E-S-R}
340: \end{eqnarray}
341: %
342: which provides a further check on the numerics.
343: %
344: As can be seen from \eref{E-S-R}, for strongly repulsive interaction the mixture behaves like a 
345: single component Fermi gas of indistinguishable particles,
346: rather than hard-core bosons \cite{BEC-BCS2,BEC-BCS3,BEC-BCS4}.
347: 
348: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
349: \begin{figure}[t]
350: \begin{center}
351: \vspace{0.5cm}
352: \includegraphics[width=0.95\linewidth]{PRL_groundstate_energies_12.eps}
353: \end{center} 
354: %
355: \caption{
356: Ground state energy in units of $\hbar^2 n^2/(2m)$ vs interaction strength $\gamma$ 
357: for different polarizations. 
358: %
359: Dashed lines compare the analytic approximations \eref{E-W}, \eref{E-S-A}, \eref{E-S-R} 
360: with the numerical solution of the integral equations (solid lines) in the 
361: different coupling regimes. 
362: %
363: The curves shown are for $a=0$, $0.2$, $0.4$, $0.6$, $0.8$, $1$ (top to bottom).
364: %
365: }
366: \label{Fig2}
367: \end{figure}
368: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
369: 
370: 
371: 
372: {\it Spin-charge separation.}
373: In the TL, the low-energy behaviour of interacting Bose and Fermi gases is
374: universal in the dispersion relation and correlation behaviour \cite{Haldane}. 
375: %
376: Spin-charge separation is one of the typical Luttinger liquid behaviours, with   
377: both the spin and charge degrees of freedom having different propagation velocities 
378: at Fermi surfaces.  
379: %
380: The charge-velocity can be calculated exactly from 
381: $v_c^2=\frac{L}{m n} \frac{\partial^2 E}{\partial L^2}$ \cite{London} 
382: for arbitrary $a$, which yields 
383: %
384: \begin{eqnarray}
385: \frac{v_c}{v_{\rm F}}=\left\{\begin{array}{ll}
386: \sqrt{\frac{a^3}{16} (1-4 b_1)+
387: a_1(1-4b_2)}, &\,\gamma \ll-1,\\
388: \sqrt{\frac{a(1-{a}/{2})\gamma}{\pi^2}+\frac{a^3}{4}+a_0},&\,|\gamma|\ll1,
389: \end{array}
390: \right.
391: \end{eqnarray}
392: %
393: where the Fermi velocity $v_{\rm F}={\hbar \pi n}/{m}$. 
394: 
395: 
396: To illustrate spin-charge separation for non-zero interaction, we show the 
397: spin and charge velocities in Fig. \ref{Fig3} as a function of $\gamma$ for $a=1$. 
398: %
399: These curves are obtained numerically from the dressed-energy 
400: formalism \cite{Takahashi, Schlottmann}. 
401: %
402: For weak interaction \cite{BEC-BCS1}, 
403: %
404: \begin{eqnarray}
405: \frac{v_{c,s}}{v_{\rm F}}&=&\frac{1}{2}\left(1\pm\frac{\gamma}{\pi^2}\right), 
406: \quad |\gamma|\ll 1,\label{smg}
407: \end{eqnarray}
408: where the upper sign refers to $v_c$. 
409: %
410: Furthermore, $\frac{v_{c}}{v_{\rm F}}=1-\frac{4\ln 2}{\gamma}$, 
411: $\frac{v_s}{v_{\rm F}}=\frac{\pi^2}{3 \gamma}$ for $\gamma \to\infty$ and 
412: $\frac{v_{c}}{v_{\rm F}}= \frac{1}{4}(1-1/\gamma)$ 
413: for $\gamma \to -\infty$ \cite{BEC-BCS1}. 
414: %
415: In the attractive regime, the charge excitations are still gapless; 
416: however, a gap opens in the spin channel \cite{Krivnov}. 
417: %
418: This gap increases with increasing $|\gamma|$, leading to a divergent spin velocity in the 
419: strongly attractive limit, with $\frac{v_s}{v_{\rm F}}=-\frac{\gamma}{ 2\sqrt2 \pi}$ as 
420: $\gamma \to -\infty$ \cite{BEC-BCS1}. 
421: %
422: Some possible ways to experimentally observe the spin-charge separation in
423: harmonically trapped atoms have been suggested \cite{SC-S}.
424: 
425: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
426: \begin{figure}[b]
427: \vspace{0.7cm}
428: \begin{center}
429: \includegraphics[width=0.80\linewidth]{chargespinvel1.eps}
430: \end{center}
431: \vspace{-1mm}
432: \caption{Spin and charge velocities vs interaction strength for $a=1$. 
433: %
434: The straight dotted lines show the lowest order in $\gamma$ from \eref{smg}, 
435: which is also accessible by field theory \cite{SC-S}.}
436: \label{Fig3}
437: \end{figure}
438: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
439: 
440: In conclusion, we have examined in detail the exactly solved two-component
441: 1D interacting Fermi gas for arbitrary polarization $a$ in the whole interaction range. 
442: %
443: Analytic results have followed for the quasimomentum density profile, the ground state energy and
444: the spin-charge velocities.
445: %
446: These quantities reveal significant insight into the nature of quantum effects in the Fermi gas
447: with arbitrary polarization.
448:  
449: \ack
450: 
451: This work has been supported by the Australian Research Council 
452: and the German Science Foundation (grant number BO/2538).
453: %
454: The authors also thank the organizers of the {\em Counting Complexity} meeting
455: in honour of Tony Guttmann's 60th birthday, where this work was presented.
456: 
457: %\clearpage
458: 
459: \section*{References}
460: 
461: \begin{thebibliography}{99}
462: 
463: \bibitem{BEC-F} 
464: Jochim S \etal 2003 {\it Science} {\bf 302}  2101
465: \nonum Chin C \etal 2004 {\it Science} {\bf 305}  1128
466: \nonum Zwierlein M W \etal 2004 \PRL {\bf 92}  120403
467: \nonum Regal C A, Greiner M and Jin D S 2004 \PRL {\bf 92} 040403
468: \nonum Bartenstein M \etal 2004 \PRL {\bf 92} 120401
469: \nonum Kinast J \etal 2005 {\it Science} {\bf 307} 1296
470: 
471: \bibitem{Fermi-1D1} 
472: Moritz H \etal 2005 \PRL {\bf 94} 210401
473: 
474: \bibitem{Fermi-1D2} 
475: Recati A, Fuchs J N and Zwerger W 2005 {\it Phys. Rev. A} {\bf 71}  033630
476: 
477: \bibitem{BEC-BCS2}
478: Astrakharchik G E, Blume D, Giorgini S and Pitaevskii L P 2004
479: \PRL {\bf 93} 050402
480: 
481: \bibitem{BEC-BCS3} 
482: Tokatly I V 2004 \PRL {\bf 93}  090405
483: 
484: \bibitem{BEC-BCS1}
485: Fuchs J N,  Recati A and Zwerger W 2004 \PRL {\bf 93}  090408
486: 
487: \bibitem{BEC-BCS4}
488: Iida T Wadati M 2005 J. Phys. Soc. Jpn. {\bf 74} 1724
489: 
490: \bibitem{BEC-BCS5} 
491: Liu X-J, Drummond P D and Hu H 2005  \PRL {\bf 94} 136406
492: 
493: \bibitem{BEC-BCS6}
494: Ortiz G and Dukelsky J 2005 {\it Phys. Rev. A} {\bf 72}  043611
495: 
496: \bibitem{Gaudin-Yang}
497: Gaudin M 1967 {\it Phys. Lett. A} {\bf 24}  55
498: \nonum Yang C N 1967 \PRL {\bf 19} 1312
499: 
500: \bibitem{SC-S}  Recati A, Fedichev P O, Zwerger W and Zoller P 2003
501: {\it J. Opt. B} {\bf 5} S55
502: 
503: \bibitem{BGM} Batchelor M T, Guan X W and McGuire J B 2004 \JPA  {\bf 37} L497
504: \nonum Batchelor M T, Guan X W, Dunning C and Links J 2005 {\it J. Phys. Soc. Jpn.} 
505: {\bf 74} S 53
506: 
507: \bibitem{Lee} Lee C 2004 \PRL {\bf 93} 120406
508: 
509: \bibitem{LM-thorem} Lieb E and Mattis D 1962 \PR {\bf 125} 164
510: 
511: \bibitem{Haldane} Haldane F D M 1981 \PRL {\bf 47} 1840
512: 
513: \bibitem{London} London F 1964 {\it Superfluids} Vol. II (New York, Dover)
514: 
515: \bibitem{Takahashi} Takahashi M 1999 {\it Thermodynamics of One-Dimensional Solvable Models}  
516: (Cambridge, Cambridge University Press).
517: 
518: \bibitem{Schlottmann} Schlottmann P 1994 {\it J. Phys.:Cond. Mat.} {\bf 6} 1359
519: 
520: \bibitem{Krivnov} 
521: Krivnov V Ya and Ovchinnikov A A 1975 {\it Sov. Phys. JETP} {\bf 40} 78
522: \nonum Musurkin I A and Ovchinnikov A A 1971 {\it Sov. Phys. Sol. St.} {\bf 12} 2031
523: 
524: \end{thebibliography}
525: 
526: \end{document}
527: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
528: 
529: