1: \documentclass[aps,pra,showpacs]{revtex4}
2: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
3: \def\rvec{\vec{r}}
4: \def\xvec{\vec{x}}
5: \def\half{{1\over 2}}
6: \makeatletter
7: %\def\@cite#1{\textsuperscript{#1}}
8: \makeatother
9: \makeatletter
10: %\def\@biblabel#1{\textsuperscript{#1}}
11: \makeatother
12: \setlength{\topmargin}{-1.5cm}
13: \setlength{\textheight}{24cm}
14: \setlength{\textwidth}{15.5cm}
15: \setlength{\oddsidemargin}{0.05cm}
16:
17: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
18: %---------------------------------------------------------------------------
19: \def\rvec{\vec{r}}
20: \def\xvec{\vec{x}}
21: \def\half{{1\over 2}}
22: \def\sla{\slash{\!\!\!} }
23: %\def\vr{\bf r}
24: %\def\vp{\bf p}
25: %\def\vR{\bf R}
26: %\def\vP{\bf P}
27: \newcommand{\eps}{\epsilon}
28: \newcommand{\vp}{\mbox{\boldmath $p$}}
29: \newcommand{\vq}{\mbox{\boldmath $q$}}
30: \renewcommand{\vr}{\mbox{\boldmath $r$}}
31: \newcommand{\vk}{\mbox{\boldmath $k$}}
32: \newcommand{\vx}{\mbox{\boldmath $x$}}
33: \newcommand{\idp}{\int \frac{{\rm d}^3\vp}{(2\pi)^3}}
34: \newcommand{\idk}{\int \frac{{\rm d}^3\vk}{(2\pi)^3}}
35: \newcommand{\ip}{\int \frac{{\rm d}^4p}{(2\pi)^4}}
36: \newcommand{\ik}{\int \frac{{\rm d}^4k}{(2\pi)^4}}
37: \newcommand{\nn}{\nonumber \\}
38: \newcommand{\tdb}{\tilde{\Delta}_b }
39: \newcommand{\tdf}{\tilde{\Delta}_f }
40: \newcommand{\tds}{\tilde{\Delta}_s }
41: \newcommand{\tdc}{\tilde{\Delta}_c }
42: \newcommand{\tdA}{\tilde{\Delta}_A }
43: \newcommand{\tdB}{\tilde{\Delta}_B }
44: \newcommand{\ua}{\uparrow}
45: \newcommand{\da}{\downarrow}
46: \newcommand{\rk}{\right)}
47: \newcommand{\lk}{\left(}
48: \newcommand{\rtk}{\right\} }
49: \newcommand{\ltk}{\left\{ }
50: \newcommand{\rdk}{\right] }
51: \newcommand{\ldk}{\left[ }
52: %
53: \newcommand{\vP}{\mbox{\boldmath $P$}}
54: \newcommand{\vR}{\mbox{\boldmath $R$}}
55: \newcommand{\bgamma}{\mbox{\boldmath $\gamma$}}
56: \newcommand{\tpsp}{\hspace{1.5em}}
57: \usepackage{graphicx}
58: %\usepackage{psfrag}
59: %---------------------------------------------------------------------------
60:
61: \begin{document}
62:
63: %%%%%%%%%%%%%
64: % Main body %
65: %%%%%%%%%%%%%
66: %
67: \title{Density waves in quasi-one-dimensional
68: atomic gas mixture of boson and two-component fermion}
69: %
70: %\date{\today}
71: %
72: \author{E. Nakano}
73: %
74: \affiliation{Yukawa Institute for Theoretical Physics, Kyoto University,
75: Sakyoku, Kyoto 606-8502, Japan}
76: %
77: \author{H. Yabu}
78: %
79: \affiliation{Department of Physics, Tokyo Metropolitan University,
80: 1-1 Minami-Ohsawa, Hachioji, Tokyo 192-0397, Japan}
81: %
82: \begin{abstract}
83: The density-wave structures are studied
84: in quasi-one-dimensional atomic gas
85: mixture of one-component bosons and two-component fermions
86: using the mean-field approximation.
87: Owing to the Peierls instability in the quasi-one-dimensional fermion system,
88: the ground state of the system shows
89: the fermion density wave and the periodic Bose-Einstein condensation
90: induced by the boson-fermion interatomic interaction.
91: For the two-component fermions,
92: two density waves appear in each component,
93: and the phase difference between them distinguishes
94: two types of ground states,
95: the in-phase and the out-phase density-waves.
96: In this paper,
97: a self-consistent method in the mean-field approximation is presented
98: to treat the density-wave states
99: in boson-fermion mixture with two-component fermions.
100: From the analysis of the effective potential and the interaction energies,
101: the density-waves are shown to appear in the ground state,
102: which are in-phase or out-phase
103: depending on the strength of the inter-fermion interaction.
104: It is also shown that the periodic Bose-Einstein condensate
105: coexists with the in-phase density-wave of fermions,
106: but, in the case of the out-phase one, only the uniform condensate appears.
107: The phase diagram of the system is given for the effective coupling constants.
108: \end{abstract}
109: %
110: \pacs{03.75Kk, 03.75.Ss, 71.45Lr, 75.30Fv}
111: %
112: \keywords{Peierls instability; periodic condensate; density wave; spin density wave}
113:
114: \maketitle
115:
116: %
117: \section{Introduction}
118: %
119:
120: Since the experimental achievement
121: of Bose-Einstein condensates (BEC) \cite{BECexp1, BECtext1}
122: and Fermi-degenerate states \cite{FDexp1},
123: the researches on the ultra-cold atomic gas have much stimulated
124: the quantum many-body physics;
125: it can create an artificial many-body system
126: with the controllable the atom number, trapping potential geometry,
127: the interatomic interactions et.al.
128:
129: In recent years,
130: the quantum-degenerate atomic gas of the boson-fermion mixture has been realized
131: with the advanced cooling technique (the sympathetic cooling)
132: \cite{BFMexp1, BFMexp2},
133: and a lot of experimental and theoretical studies
134: have been done for it
135: \cite{BFMtheory1,BFMtheory2,BFMtheory3,BFMtheory4,BFMtheory5,BFMtheory6,BFMtheory7,BFMtheory8,BFMtheory9}.
136:
137: The quasi-one-dimensional atomic (Q1D) gas of the boson-fermion mixture
138: should open up many interesting possibilities in the quantum many-body system
139: \cite{MYS1}:
140: the Tonk-Girardeau gas \cite{tonk1, tonk2},
141: and other regimes of trapped 1D bosons \cite{regime1} and
142: of fermions \cite{regime2}.
143: As discussed in \cite{MYS1},
144: the Q1D atomic gas can be realized, for example,
145: in the axially-deformed harmonic oscillator (HO) trap
146: with the axial and radial frequencies $\omega_{a,r}$,
147: which satisfy $\omega_a \ll \omega_r$ and
148: should be larger than the interatomic interaction energies and
149: the healing length of the system \cite{MYS1, Das1, Pita1}.
150: Experimentally, the quasi-one-dimensional systems has already been realized
151: in the boson atomic gas \cite{BFMexp1, BEClow1, BEClow2, BEClow3},
152: where the BEC solitons propagating
153: in the axial direction have been observed \cite{soliton1, soliton2, soliton3}.
154:
155: A particular property of the fermionic Q1D system is
156: the occurrence of the density wave
157: due to the Peierls instability
158: with the frequency of twice the Fermi wave number $2 k_F$.
159: In \cite{MYS1},
160: we investigated the ground-state properties and
161: the collective excitations
162: in the mixture of one-component boson and one-component fermion,
163: and showed that the fermion density wave appears in the ground state of the system,
164: which is similar with the charge density wave (CDW)
165: in the Q1D electron system \cite{peierl, Kittel, kagoshima},
166: and we also found that
167: the BEC of the bosons becomes spatially periodic (periodic BEC).
168:
169: In the present paper,
170: we investigate the ground state properties
171: of the Q1D atomic gas at zero-temperature ($T=0$)
172: of one-component boson and two-component fermion,
173: which interact through the atomic collisions.
174: In this system,
175: two density waves can appear in each component of fermions
176: in the ground state,
177: and two types of the ground-state structure
178: are distinguished with the relative phase between them,
179: the in-phase and the out-phase density-waves (Fig.~1);
180: %------------------------------------------------------
181: \begin{figure}
182: \begin{center}
183: \includegraphics[height=6cm]{Fig1.eps}
184: \end{center}
185: \vspace{-1.5cm}
186: \caption{In-phase and out-phase density waves
187: in fermion densities $\rho_\uparrow(z)$ (solid curve)
188: and $\rho_\downarrow(z)$ (dashed curve).}
189: \label{Fig1}
190: \end{figure}
191: %------------------------------------------------------
192: the latter corresponds to the spin density wave (SDW)
193: in the electron system \cite{kagoshima}.
194: The density-wave structure in the ground state depends on
195: the strength of the interatomic interactions;
196: especially important is the inter-component interaction of fermions,
197: which is absent in the previous study \cite{MYS1}.
198: Also, the structure of BEC, uniform or periodic,
199: should be influenced by the fermion density-wave structure.
200:
201: In Sec.~$\mbox{I\hspace{-.1em}I}$,
202: we introduce a model Hamiltonian for the system,
203: and propose a new method to treat the periodic order parameters
204: for boson and fermion-density fields,
205: which are proved to be self-consistent
206: in the mean-field approximation, and
207: construct the effective potential with this method.
208: In Sec.~$\mbox{I\hspace{-.1em}I\hspace{-.1em}I}$,
209: the density-wave structures in the ground state
210: are classified by analytical evaluation of the effective potential
211: for the cases of the repulsive/attractive fermion-fermion interactions;
212: the results are summarized in the phase diagram
213: for the effective coupling constants of the interatomic interactions.
214: The final section is devoted to the summary of the paper.
215:
216: %
217: \section{Formulation}
218: %
219:
220: \subsection{Hamiltonian}
221:
222: We consider the Q1D system of atomic gas mixture
223: of one-component boson and two-component fermion at $T=0$.
224: In this paper,
225: we call two components of fermion ``up'' and ``down'', and
226: denote them with the symbol $\uparrow$ and $\downarrow$.
227: It should be noted that they should not only be the up and down components of spin 1/2 atom,
228: but any two of hyperfine levels or different atomic species.
229:
230: The masses of the bosons and the fermions are denoted
231: by $m_b$ and $m_f$.
232: In this paper,
233: we consider the case of degenerate two-component fermion for simplicity,
234: and assume that the up and down fermions have the same mass
235: (also, we assume that the other properties, the interaction strengths et.al.,
236: are degenerate).
237:
238: The boson-fermion mixture is assumed to be trapped in the HO potentials
239: with the same frequencies:
240: %
241: \begin{equation}
242: U_{b,f}=m_{b,f}\ltk \omega_r^2(x^2+y^2)+\omega_a^2z^2\rtk/2,
243: \label{HOP}
244: \end{equation}
245: %
246: If the HO potentials are enough long in the z-direction,
247: $\omega_r \gg \omega_a$,
248: the system becomes Q1D.
249: (For the other conditions in the realization of the Q1D system,
250: see \cite{MYS1,Das1,Pita1}.)
251:
252: In the Q1D system,
253: the dynamical degrees of freedom of atoms are represented
254: by the 1D field operators,
255: $\phi(z)$ for boson and $\psi_{\uparrow, \downarrow}(z)$ for fermions.
256: Using these field,
257: the Hamiltonian of the system is given by
258: %
259: \begin{eqnarray}
260: {\rm H} &=& \int dz \phi^\dagger(z)
261: \left[ -\frac{\hbar^2}{2m_b}
262: \frac{{\rm d}^2}{{\rm d}z^2}
263: +U_b'(z) -\mu_b
264: \right] \phi(z) \nn
265: %
266: &&+ \int dz \sum_{s=\uparrow, \downarrow}
267: \psi_s^\dagger(z)
268: \left[ -\frac{\hbar^2}{2m_f}
269: \frac{{\rm d}^2}{{\rm d}z^2}
270: +U_f'(z) -\mu_f \right] \psi_s(z) \nn
271: %
272: &&+ \int dz \phi^\dagger(z)
273: \left[ \frac{g_{bb}}{2} \phi^\dagger(z) \phi(z)
274: + g_{bf} \sum_{s=\uparrow, \downarrow}
275: \psi_s^\dagger(z) \psi_s(z)
276: \right] \phi(z) \nn
277: %
278: &&+ \int dz\; g_{ff} \psi_\uparrow^\dagger(z)
279: \psi_\uparrow(z)
280: \psi_\downarrow^\dagger(z)
281: \psi_\downarrow(z),
282: \label{Q1DH}
283: \end{eqnarray}
284: %
285: where $\mu_{b,f}$ are one-dimensional chemical potentials
286: for the bosons and the fermions.
287: The coupling constants $g_{bb,bf,ff}$ in eq.~(\ref{Q1DH})
288: represent the strengths of the boson-boson, boson-fermion
289: and fermion-fermion interactions, respectively.
290: In the case of the large attractive boson-boson interaction ($g_{bb} \ll 0$),
291: the BEC instability appears
292: and competes with the Peierls instability.
293: In the present paper,
294: to concentrate on the density wave states,
295: we consider the case of the repulsive boson-boson interaction ($g_{bb}>0$) only.
296: For the Q1D atomic gas,
297: the coupling constants are determined by the (3D) s-wave scattering lengths
298: $a_{bb,bf,ff}$ of the atomic collisions \cite{pethick}:
299: $g_{bb}=2\hbar \omega_r a_{bb}$,
300: $g_{bf}=2\hbar \omega_r a_{bf}$, and
301: $g_{ff}=2\hbar \omega_r a_{ff}$,
302: where $\omega_r$ is the radial frequency of the (3D) HO potential (\ref{HOP}).
303: The $U_{b,f}'(z)$ in eq.~(\ref{Q1DH}) are the $z$-directional part
304: in the original HO potential (\ref{HOP}),
305: which are neglected in the present calculation.
306: (About the effects of the HO potential, see \cite{MYS1}.)
307:
308: \subsection{Density-wave ground state}
309:
310: It is well known in the Q1D fermion system \cite{kagoshima,peierl,MYS1,Kittel}
311: and also in the boson-fermion mixture \cite{MYS1}
312: that the homogeneous state is unstable (Peierls instability)
313: for the formation of the density wave with the wave number $2k_F$.
314: To describe the density-wave states,
315: we introduce the $2k_F$-periodic order parameters
316: for the boson and fermion fields:
317: %
318: \begin{eqnarray}
319: \langle \phi(z) \rangle
320: &\equiv& \Phi(z) = b_0+2b_c \cos(2k_F z),
321: \label{BD1} \\
322: %
323: \rho_\downarrow(r)
324: &\equiv& \langle \psi_\downarrow^\dagger(z)
325: \psi_\downarrow(z) \rangle
326: %
327: =f_0 +2f_c \cos(2k_F z+\varphi)
328: +2f_b \cos(2k_F z),
329: \label{FD1} \\
330: %
331: \rho_\uparrow(r)
332: &\equiv& \langle \psi_\uparrow^\dagger(z)
333: \psi_\uparrow(z) \rangle
334: %
335: =f_0 +2f_c \cos(2k_F z +\varphi+\theta)
336: +2f_b \cos(2k_F z),
337: \label{FD2}
338: \end{eqnarray}
339: %
340: where $\varphi$ is the phase-difference angles
341: between boson- and fermion-density waves,
342: and $\theta$ is that between the up- and down-fermion density waves;
343: $\theta=0 ~(\pi)$ corresponds to the in-phase (out-phase) density wave of fermions (Fig.~1).
344: The $b_c$ and $f_c$ in eq.~(\ref{FD2}) are the amplitudes of the boson
345: and fermion density waves,
346: and the third components with the amplitude $f_b$ in $\rho_{\downarrow,\uparrow}$
347: have to be added to reflect the fermion-density waves
348: induced by the periodic BEC
349: (the term with the amplitude $b_c$ in the boson order parameter).
350: Thus, the amplitude $f_b$ will be shown
351: to be proportional to the $b_c$ at the end of this section.
352: The constants $b_0$ and $f_0$ represent the homogeneous terms
353: in the boson and fermion densities,
354: which are determined by the boson and fermion average densities,
355: $n_b$ and $n_f$ after the spatial integration of the local densities:
356: %
357: \begin{equation}
358: n_b=b_0^2+2b_c^2, \quad
359: n_f=n_{f, \uparrow}=n_{f, \downarrow}=f_0.
360: \label{density}
361: \end{equation}
362: %
363: In the present study,
364: we do not consider the case
365: where the spin-polarized states
366: with $n_{f, \uparrow}-n_{f, \downarrow} \neq 0$
367: become stable;
368: in the cases of the very large fermion density
369: (or the very strong fermion-fermion interaction),
370: such a polarized state might be stable
371: (the collective ferromagnetic states) \cite{SpinPol}.
372:
373: Using eqs.~(\ref{BD1}-\ref{FD2}) for the Hamiltonian (\ref{Q1DH}),
374: we obtain the mean-field Hamiltonian density:
375: %
376: \begin{eqnarray}
377: {\mathcal H}_{MF}
378: ={\mathcal H}_F
379: +{\mathcal H}_B,
380: \label{MFH0}
381: \end{eqnarray}
382: %
383: where the fermion part ${\mathcal H}_F$
384: includes the contributions from the fermion kinetic term
385: and the fermion-fermion and boson-fermion interaction terms,
386: and the boson part ${\mathcal H}_B$
387: comes from the boson kinetic and boson-boson interaction terms.
388: Using the periodic regularization with the spatial length $L$
389: and taking the limit $L\rightarrow \infty$,
390: the fermion part for the momentum $-2k_F < k < 0$,
391: denoted by ${\mathcal H}_F^-$, becomes
392: %
393: \begin{eqnarray}
394: {\mathcal H}_F^-&=& \int \frac{dk}{2\pi}
395: \left( \begin{array}{c}
396: \psi_\uparrow(k) \\
397: \psi_\uparrow(k+2k_F)
398: \end{array}
399: \right)^\dagger
400: %
401: \left[ \begin{array}{c c}
402: \epsilon(k) & \Delta_b +\Delta_b' +e^{i\varphi} \Delta_f \\
403: \Delta_b^* +{\Delta_b'}^* +e^{-i\varphi} \Delta_f^* & \epsilon(k+2k_F)
404: \end{array} \right]
405: %
406: \left( \begin{array}{c}
407: \psi_\uparrow(k) \\
408: \psi_\uparrow(k+2k_F)
409: \end{array}\right) \nn
410: %
411: &+& \int \frac{dk}{2\pi}
412: \left( \begin{array}{c}
413: \psi_\downarrow \\
414: \psi_\downarrow(k+2k_F)
415: \end{array} \right)^\dagger
416: %
417: \left[\begin{array}{c c}
418: \epsilon(k) & \Delta_b +\Delta_b' +e^{i(\theta+\varphi)} \Delta_f \\
419: \Delta_b^* +{\Delta_b'}^* +e^{-i(\theta+\varphi)} \Delta_f^* & \epsilon(k+2k_F)
420: \end{array} \right]
421: %
422: \left( \begin{array}{c}
423: \psi_\downarrow(k) \\
424: \psi_\downarrow(k+2k_F)
425: \end{array} \right) \nn
426: %
427: &-& \frac{g_{ff}}{2} f_0^2
428: -\frac{1}{g_{ff}}
429: \left[ \Delta_f^2\cos(\theta)
430: +2\Delta_f \Delta_b'
431: \cos\left(\frac{\theta}{2} \right)
432: \cos\left(\varphi+\frac{\theta}{2} \right)
433: +\Delta_b'^2 \right].
434: \label{MFH1}
435: \end{eqnarray}
436: %
437: The fermion part, ${\mathcal H}_F^+$,
438: corresponding to the positive momentum $0<k<2k_F$,
439: is obtained
440: by the replacement $k_F \rightarrow -k_F$ in eq.~(\ref{MFH1}),
441: and the total fermion part becomes
442: ${\mathcal H}_F={\mathcal H}_F^-+ {\mathcal H}_F^+$.
443: The $\epsilon(k)$ is the single-particle energy
444: %
445: $\epsilon(k) = \hbar^2 k^2/2 m_f-\mu_f +g_{ff} f_0 +g_{bf}n_b$,
446: %
447: and the gap functions $\Delta$ are defined by
448: %
449: \begin{equation}
450: \Delta_b \equiv 2b_0 b_c g_{bf}, \quad
451: \Delta_f \equiv f_c g_{ff}, \quad
452: \Delta_b' \equiv f_b g_{ff}.
453: \label{gapf}
454: \end{equation}
455:
456: Using the similar technique with the Bogoliubov transformation,
457: the ${\mathcal H}_F$ is diagonalized
458: (for the detailed calculation, see Appendix A):
459: %
460: \begin{eqnarray}
461: {\mathcal H}_F^- &=& \int \frac{dk}{2\pi}
462: \sum_{s={\uparrow, \downarrow}}
463: \left[ E^-_s(k) {\alpha^-_s}^\dagger(k)
464: \alpha^-_s(k)
465: +E^+_s(k) {\alpha^+_s}^\dagger(k)
466: \alpha^+_s(k) \right] \nn
467: %
468: &-& \frac{g_{ff}}{2} f_0^2
469: -\frac{1}{g_{ff}}
470: \left[ \Delta_f^2\cos(\theta)
471: +2\Delta_f \Delta_b'
472: \cos\left(\frac{\theta}{2} \right)
473: \cos\left(\varphi+\frac{\theta}{2} \right)
474: +\Delta_b'^2 \right],
475: \label{fermi1}
476: \end{eqnarray}
477: %
478: where $\alpha^\pm_{\uparrow, \downarrow}(k)$ are the annihilation operators
479: for the quasi-particle with the momentum $-2k_F < k < 0$
480: in the energies:
481: %
482: \begin{eqnarray}
483: E^\pm_\uparrow(k) &=& \left[ \epsilon(k)+v_F(k+k_F) \right]
484: \pm \sqrt{ v_F^2(k+k_F)^2
485: +|\Delta_{\uparrow}|^2 },
486: \label{qpe1} \\
487: %
488: E^\pm_\downarrow(k) &=& \left[ \epsilon(k)+v_F(k+k_F) \right]
489: \pm \sqrt{ v_F^2(k+k_F)^2
490: +|\Delta_{\downarrow}|^2},
491: \label{qpe2}
492: \end{eqnarray}
493: %
494: where $v_F$ is the Fermi velocity.
495: The quasi-particle energies for the momentum $0<k<2k_F$ are obtained by replacing
496: $k_F, v_F \rightarrow -k_F, -v_F$ in eqs.~(\ref{qpe1},\ref{qpe2}).
497: The effective gap functions $\Delta_{\uparrow, \downarrow}$ in eqs.~(\ref{qpe1},\ref{qpe2})
498: are defined by
499: %
500: \begin{eqnarray}
501: |\Delta_{\uparrow}|^2
502: &\equiv& \left| \Delta_b +\Delta_b'
503: + e^{i\varphi}\Delta_f \right|^2
504: %
505: =(\Delta_b +\Delta_b')^2
506: +2 (\Delta_b+\Delta_b') \Delta_f \cos(\varphi)
507: +\Delta_f^2,
508: \label{egf1}\\
509: %
510: |\Delta_{\downarrow}|^2
511: &\equiv& \left| \Delta_b +\Delta_b'
512: + e^{i(\varphi+\theta)}\Delta_f \right|^2
513: %
514: =(\Delta_b+\Delta_b')^2
515: +2 (\Delta_b +\Delta_b') \Delta_f \cos(\theta+\varphi)
516: +\Delta_f^2.
517: \label{egf2}
518: \end{eqnarray}
519:
520: In Fig.~\ref{level1}, we show the quasi-particle energy in eqs.~(\ref{qpe1},\ref{qpe2}).
521: It should be noted that the energy gaps appear at $k =\pm k_F$
522: in the quasi-particle spectrum.
523:
524: %------------------------------------------------------
525: \begin{figure}
526: \begin{center}
527: \includegraphics[height=5cm]{Quasi-Ene1.eps}
528: \end{center}
529: \caption{Illustration of the quasi-particle spectrum,
530: $E_{\uparrow, \downarrow}^{(\pm)}(k)$.
531: We scaled quantities by the fermi energy and fermi momentum,
532: and set $\Delta_{\uparrow, \downarrow}/2\epsilon_F=0.1$. }
533: \label{level1}
534: \end{figure}
535: %------------------------------------------------------
536:
537: In the density-wave ground state,
538: the up- or down-fermions occupy
539: the lower branch of the quasi-particle spectrum
540: so that the energy density of fermions in this state
541: is given by the summation of $E^-_{\uparrow (\downarrow)}(k)$
542: in eqs.~(\ref{qpe1},\ref{qpe2})
543: up to the Fermi momentum $k_F$:
544: %
545: \begin{eqnarray}
546: \langle {\mathcal H}_F\rangle
547: &&= \sum_{s=\uparrow, \downarrow}
548: \int \frac{dk}{2\pi}
549: \left[ E^-_s(k,-k_F) \theta(k_F-k)\theta(k)
550: +E^-_s(k,k_F) \theta(k_F+k)\theta(-k) \right] \nn
551: &&\quad
552: - g_{ff} f_0^2
553: -\frac{2}{g_{ff}} \left[ \Delta_f^2\cos(\theta)
554: +2\Delta_f \Delta_b'
555: \cos\left(\frac{\theta}{2} \right)
556: \cos\left(\varphi+\frac{\theta}{2} \right)
557: +\Delta_b'^2 \right] \nn
558: %
559: &&= n_f \left[ \frac{4\epsilon_F}{3}
560: +g_{ff} f_0
561: -\mu_f
562: + g_{bf} n_b \right] \nn
563: &&\quad
564: -\frac{n_f}{2} \sum_{s=\uparrow, \downarrow}
565: \left[ \sqrt{4\epsilon_F^2 +\Delta_s^2}
566: +\frac{\Delta_s^2}{2 \epsilon_F}
567: \log{\frac{2 \epsilon_F
568: +\sqrt{4 \epsilon_F^2
569: +\Delta_s^2}}{\Delta_s}} \right] \nn
570: %
571: &&\quad
572: - g_{ff}f_0^2
573: -\frac{2}{g_{ff}}
574: \left[ \Delta_f^2\cos(\theta)
575: +2\Delta_f \Delta_b'
576: \cos\left(\frac{\theta}{2} \right)
577: \cos\left(\varphi+\frac{\theta}{2} \right)
578: +\Delta_b'^2 \right],
579: \label{fmfe}
580: \end{eqnarray}
581: where we have used
582: $\langle {\mathcal H}_F\rangle=
583: \langle {\mathcal H}_F^-\rangle+ \langle{\mathcal H}_F^+\rangle
584: =2 \langle{\mathcal H}_F^-\rangle$.
585:
586: Now we turn to the evaluation of the mean-field energy
587: for the boson part ${\mathcal H}_B$ in eq.~(\ref{MFH1}).
588: Substituting eq.~(\ref{BD1}) into the boson part,
589: we obtain the boson energy density up to $O(b_c^2)$:
590: %
591: \begin{eqnarray}
592: \langle {\mathcal H}_B \rangle
593: &=& 2 \left[ \epsilon_b(2k_F)
594: +6 \frac{g_{bb}}{2}b_0^2 \right] b_c^2
595: %
596: +\frac{g_{bb}}{2}b_0^4
597: +6 \frac{g_{bb}}{2} b_c^4 \nn
598: %
599: &\simeq& 2 \left[ \epsilon_b(2k_F)
600: +5 \frac{g_{bb}}{2}b_0^2 \right]b_c^2
601: %
602: +\frac{g_{bb}}{2} b_0^2 n_b \nn
603: %
604: &=& 2 \epsilon_b(2k_F) b_c^2
605: +10 \frac{g_{bb}}{2} \frac{\Delta_b^2}{4g_{bf}^2}
606: +\frac{g_{bb}}{2} b_0^2 n_b \nn
607: %
608: &\simeq& \frac{g_{bb}}{2} n_b^2
609: +\left( \frac{k_F^2}{m_b n_b} +g_{bb} \right)
610: \left( \frac{\Delta_b}{g_{bf}} \right)^2.
611: \label{bmfe}
612: \end{eqnarray}
613: %
614: In the derivation of the last line,
615: we have used the solution of
616: $n_b=b_0^2+2b_c^2$ (the average boson density)
617: and $\Delta_b \equiv 2b_0 b_c g_{bf}$:
618: %
619: \begin{eqnarray}
620: b_0^2 &=& \frac{1}{2}
621: \left[ n_b
622: +\sqrt{n_b^2 -2 \left(\frac{\Delta_b}{g_{bf}} \right)^2}
623: \right], \\
624: %
625: 2b_c^2 &=& \frac{1}{2}
626: \left[ n_b
627: -\sqrt{n_b^2 -2 \left(\frac{\Delta_b}{g_{bf}} \right)^2}
628: \right], \\
629: %
630: n_b^2 &\ge& 2 \left(\frac{\Delta_b}{g_{bf}} \right)^2,
631: \end{eqnarray}
632: %
633: where we have assumed the small modulation
634: in the boson density $b_0^2 \gg 2 b_c^2$,
635: and taken the solution satisfying $b_0^2=n_b$ and $2b_c^2=0$ at $\Delta_b=0$.
636:
637: Thus the total energy density is given by the sum of eqs.~(\ref{fmfe},\ref{bmfe}) :
638: %
639: \begin{eqnarray}
640: \mathcal{E}_{tot}(\Delta_b, \Delta_f, \Delta_b'; \varphi, \theta)
641: &\equiv& \langle {\mathcal H}_{MF}\rangle
642: =\langle {\mathcal H}_F\rangle
643: +\langle {\mathcal H}_B\rangle.
644: \label{EVH}
645: \end{eqnarray}
646: %
647: Explicit representation of the energy difference
648: $\tilde{\mathcal{E}}_{tot}(\Delta_b, \Delta_f, \Delta_b')$
649: between the density-wave state and the normal state
650: ($\Delta_b =\Delta_f =\Delta_b =0$)
651: is given by
652: %
653: \begin{eqnarray}
654: && \tilde{\mathcal{E}}_{tot}(\Delta_b, \Delta_f, \Delta_b')
655: \equiv \mathcal{E}_{tot}(\Delta_b, \Delta_f, \Delta_b')
656: -\mathcal{E}_{tot}(0, 0, 0) \nn
657: %
658: &&\qquad
659: = -\sum_{s=\uparrow, \downarrow}
660: \frac{n_f}{2}
661: \left[ \sqrt{4\epsilon_F^2 +\Delta_s^2}
662: +\frac{\Delta_s^2}{2 \epsilon_F}
663: \log \frac{2 \epsilon_F
664: +\sqrt{4 \epsilon_F^2
665: +\Delta_s^2}
666: }{\Delta_s}
667: \right]
668: %
669: +\left( \frac{k_F^2}{m_b n_b} +g_{bb} \right)
670: \left(\frac{\Delta_b}{g_{bf}} \right)^2 \nn
671: %
672: &&\qquad\quad
673: +2 n_f \epsilon_F
674: -\frac{2}{g_{ff}}
675: \left[ \Delta_f^2\cos\theta
676: +2 \Delta_f \Delta_b'
677: \cos\left(\frac{\theta}{2} \right)
678: \cos\left(\varphi+\frac{\theta}{2} \right)
679: +\Delta_b'^2 \right]
680: \label{ED1} \\
681: %
682: &&\qquad
683: \simeq 2 n_f \epsilon_F
684: \Bigg\{ -\sum_{s=\uparrow, \downarrow}
685: \frac{1}{4}
686: \left[ 1
687: -2\log \left( \frac{\Delta_s
688: }{4\epsilon_F}
689: \right)
690: \right]
691: \left( \frac{\Delta_s}{2\epsilon_F} \right)^2
692: +\frac{1}{\zeta}
693: \left( \frac{\Delta_b}{2\epsilon_F} \right)^2
694: \nn
695: %
696: &&\qquad\qquad\qquad
697: -\frac{2}{\xi}
698: \left[ \left( \frac{\Delta_f}{2\epsilon_F} \right)^2
699: \cos\theta
700: +2 \frac{\Delta_f \Delta_b'}{(2\epsilon_F)^2}
701: \cos\left(\frac{\theta}{2} \right)
702: \cos\left(\varphi+\frac{\theta}{2} \right)
703: +\left(\frac{\Delta_b'}{2\epsilon_F}\right)^2
704: \right] \Bigg\},
705: \label{ED3}
706: \end{eqnarray}
707: %
708: where dimensionless parameters $\zeta$ and $\xi$
709: (effective coupling constants) have been introduced:
710: %
711: \begin{eqnarray}
712: \zeta &\equiv& \frac{g_{bf}^2}{\pi \hbar g_{bb} v_F}
713: \left[ 1 +\left( \frac{v_F m_f}{v_B m_b} \right)^2 \right]^{-1},
714: \label{vzeta}\\
715: %
716: \xi &\equiv& \frac{g_{ff}}{\pi \hbar v_F},
717: \label{vxi}
718: \end{eqnarray}
719: %
720: and $v_B=\sqrt{g_{bb} n_b/m_b}$ is the bosonic velocity.
721:
722: The stable ground state of the system is determined
723: from the minimum point of eq.~(\ref{ED3}).
724:
725: \subsection{Consistency of the order parameters in mean-field approximation}
726:
727: Here we check the consistency
728: of the Fermion density ansatz~(\ref{FD1}, \ref{FD2})
729: in the mean-field approximation,
730: where the fermion densities $\rho_{\uparrow, \downarrow}(z)$
731: are represented by
732: %
733: \begin{eqnarray}
734: \rho_s(z) &\equiv& \langle \psi^\dagger_s(z)\psi_s(z) \rangle \nn
735: &=& \int \frac{dp}{2\pi} \int \frac{dk}{2\pi}
736: \langle \psi^\dagger_s(p)\psi_s(k) \rangle
737: e^{i(p-k)z} \nn
738: %
739: &=& \int \frac{dk}{2\pi}
740: \langle \psi^\dagger_s(k) \psi_s(k) \rangle \nn
741: %
742: && +\int \frac{dk}{2\pi}
743: \langle \psi^\dagger_s(k+2k_F) \psi_s(k) \rangle
744: e^{i2k_F z}
745: +{\rm C.C.} , \qquad
746: (s =\uparrow, \downarrow).
747: \label{dex}
748: \end{eqnarray}
749: %
750: In the quasi-particle representation (shown in Appendix A),
751: the expectation values appeared in eq.~(\ref{dex}) become
752: %
753: \begin{eqnarray}
754: \int \frac{dk}{2\pi} \langle \psi^\dagger_s(k)\psi_s(k) \rangle
755: &=& n_f,
756: \label{dexqp1}\\
757: %
758: \int \frac{dk}{2\pi} \langle \psi^\dagger_s(k+2k_F) \psi_s(k) \rangle
759: &=& 2 \int_{-k_F}^0 \frac{dk}{2\pi}
760: \frac{-\Delta_s}{2 \sqrt{ \ltk \epsilon(k)
761: -\epsilon(k+2k_F) \rtk^2
762: +|\Delta_s|^2 }}.
763: \label{dexqp2}
764: \end{eqnarray}
765: %
766: Using the above equations in eq.~(\ref{dex}),
767: we obtain the quasi-particle representation of the fermion densities:
768: %
769: \begin{eqnarray}
770: \rho_\uparrow(z) = n_f
771: &+& 2 \int_{-k_F}^0 \frac{dk}{2\pi}
772: \frac{-(\Delta_b+\Delta_b')}{\sqrt{ \ltk\epsilon(k)
773: -\epsilon(k+2k_F)\rtk^2
774: +|\Delta_\uparrow|^2 }}
775: \cos(2k_F z) \nn
776: %
777: &+& 2 \int_{-k_F}^0 \frac{dk}{2\pi}
778: \frac{-\Delta_f}{\sqrt{ \ltk \epsilon(k)
779: -\epsilon(k+2k_F) \rtk^2
780: +|\Delta_\uparrow|^2 }}
781: \cos(2k_F z+\varphi),
782: \label{FD1re} \\
783: %
784: {} \nn
785: %
786: \rho_\downarrow(z) = n_f
787: &+& 2 \int_{-k_F}^0 \frac{dk}{2\pi}
788: \frac{-(\Delta_b +\Delta_b')}{\sqrt{ \ltk\epsilon(k)
789: -\epsilon(k+2k_F)\rtk^2
790: +|\Delta_\downarrow|^2 }}
791: \cos(2k_F z) \nn
792: %
793: &+& 2 \int_{-k_F}^0 \frac{dk}{2\pi}
794: \frac{-\Delta_f}{\sqrt{ \ltk\epsilon(k)
795: -\epsilon(k+2k_F)\rtk^2
796: +|\Delta_\downarrow|^2 }}
797: \cos(2k_F z+\theta+\varphi).
798: \label{FD2re}
799: \end{eqnarray}
800: %
801: Here, the densities $\rho_{\uparrow, \downarrow}(z)$ above show
802: the same $z$-dependence with those in eqs.~(\ref{FD1}, \ref{FD2});
803: it proved the consistency of the present method.
804:
805: In the present paper,
806: we consider the case of $n_{f,\uparrow}=n_{f,\downarrow}$, i.e., no polarization,
807: and have assumed the same amplitudes in ansatz (\ref{FD1},\ref{FD2}),
808: so that we can put $|\Delta_{\downarrow}|=|\Delta_{\uparrow}|$.
809: From eqs.~(\ref{egf1}, \ref{egf2})
810: for the gap functions $\Delta_{\uparrow, \downarrow}$,
811: we obtain the relation between the phase-difference angles:
812: $\cos(\varphi)=\cos(\theta+\varphi)$.
813:
814: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
815: %Since, in eq.~(\ref{density}),
816: %the same densities are presumed for the up- and down-fermions,
817: %the gap parameters of the up- and down-fermions
818: %have the same amplitudes $\Delta_{\downarrow}=\Delta_{\uparrow}$,
819: %and it leads to the relation between density-wave phase angles
820: %$\cos(\varphi)=\cos(\theta+\varphi)$.
821: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
822:
823: Furthermore,
824: we can derive the relation among the gap functions
825: at the stationary points of the mean-field energy (\ref{EVH}).
826: From the gap equations derived in Appendix~B,
827: we obtain
828: %
829: \begin{equation}
830: \Delta_b'=-\frac{\xi}{2 \zeta} \Delta_b
831: -\Delta_f \cos\left(\frac{\theta}{2} \right)
832: \cos\left(\varphi+\frac{\theta}{2} \right).
833: \label{delbp0}
834: \end{equation}
835: %
836: Using the definition of the gap functions in (\ref{gapf}),
837: it can be written as the relation of the density-wave amplitudes:
838: %
839: \begin{equation}
840: f_b = -\frac{\Delta_b}{2 G_{bf}}
841: -f_c \cos\left(\frac{\theta}{2} \right)
842: \cos\left(\varphi+\frac{\theta}{2} \right),
843: \label{delbp}
844: \end{equation}
845: %
846: where the effective boson-fermion interaction parameter $G_{bf}$ is defined by
847: %
848: \begin{equation}
849: G_{bf} \equiv g_{bf}^{2} \left( \frac{k_F^2}{m_b n_b} +g_{bb} \right)^{-1}.
850: \label{effbf}
851: \end{equation}
852: %
853: In the present paper, we consider the case of $g_{bb}>0$,
854: so that the parameter $G_{bf}$ is positive ($G_{bf}>0$).
855: The relation (\ref{delbp0}) shows
856: that $\Delta_b'$ is related to
857: a linear combination of $\Delta_b$ and $\Delta_f$ in the ground state.
858: In the next section,
859: we will show that $\Delta_b'$ is proportional only
860: to the boson amplitude $\Delta_b$
861: after fixing the phase-difference angles in the ground state.
862:
863: \subsection{Interaction energy and phase differences}
864:
865: To determine the ground state,
866: we should fix the phase-difference angles $\varphi$ and $\theta$
867: so as to gain the largest interaction energy for the fixed gap functions.
868: For this purpose,
869: we reevaluate the $\varphi$- and $\theta$-dependent terms in (\ref{ED3}),
870: included in the boson-fermion and fermion-fermion interaction terms in eq.~(\ref{Q1DH}),
871: in the mean-field approximation.
872:
873: Using the parametrization for order parameters (\ref{BD1},\ref{FD1},\ref{FD2}),
874: the boson-fermion interaction term becomes
875: %
876: \begin{eqnarray}
877: && g_{bf} \int dz \sum_{s=\uparrow, \downarrow}
878: \phi^\dagger(z) \phi(z) \psi_s^\dagger(z) \psi_s(z) \nn
879: %
880: &&\Rightarrow g_{bf} \int_0^L \frac{dz}{L}
881: \sum_s \Phi^\dagger(z) \Phi(z)
882: \langle \psi_s^\dagger(z) \psi_s(z) \rangle |_{L\rightarrow \infty} \nn
883: %
884: &&\qquad= 2 g_{bf}
885: \left[ n_b f_0
886: +4 b_0 b_c f_c
887: \cos\left(\frac{\theta}{2} \right)
888: \cos\left(\varphi+\frac{\theta}{2} \right)
889: +4b_0 b_c f_b \right] \\
890: %
891: &&\qquad= 2 g_{bf} n_b n_f -2\frac{\Delta_b^2}{G_{bf}},
892: \label{bfie}
893: \end{eqnarray}
894: %
895: where, in the derivation of the last line,
896: we have used the relation (\ref{delbp}),
897: which is valid for the ground state.
898: Eq.~(\ref{bfie}) shows that the boson-fermion interaction term
899: is independent of the phase-difference angles in the ground state,
900: and plays no role in the phase difference determination.
901:
902: In the same way,
903: we reevaluate
904: the fermion-fermion interaction term:
905: %
906: \begin{eqnarray}
907: && g_{ff}\int dz\; \psi_\uparrow^\dagger(z)
908: \psi_\uparrow(z)
909: \psi_\downarrow^\dagger(z)
910: \psi_\downarrow(z) \nn
911: %
912: &&\Rightarrow g_{ff} \int_0^L \frac{dz}{L}
913: \langle \psi_\uparrow^\dagger(z) \psi_\uparrow(z) \rangle
914: \langle \psi_\downarrow^\dagger(z) \psi_\downarrow(z) \rangle |_{L\rightarrow \infty} \nn
915: %
916: &&\qquad= 2 g_{ff} \ldk \frac{f_0^2}{2}
917: +\ltk f_c \cos\left(\frac{\theta}{2} \right)
918: \cos\left(\varphi+\frac{\theta}{2} \right)
919: +f_b \rtk^2 \right. \nn
920: %
921: &&\qquad\qquad\qquad
922: + f_c^2 \left. \ltk -\cos\left(\frac{\theta}{2} \right)^2
923: \cos\left(\varphi+\frac{\theta}{2} \right)^2
924: +\cos\theta \rtk \rdk \nn
925: %
926: &&\qquad= 2 g_{ff}\ldk \frac{n_f^2}{2}
927: +\frac{\Delta_b^2}{4G_{bf}^2} \right. \nn
928: %
929: &&\qquad\qquad\qquad
930: +f_c^2 \left. \ltk \cos\left(\frac{\theta}{2} \right)^2
931: \sin\left(\varphi+\frac{\theta}{2} \right)^2
932: -\sin\left(\frac{\theta}{2} \right)^2 \rtk \rdk.
933: \label{ffie}
934: \end{eqnarray}
935: %
936: The phase-difference angles $\varphi$ and $\theta$ should be determined
937: to make the energy gain largest in eq.~(\ref{ffie})
938: for fixed values of the coupling constant $g_{ff}$.
939: Therein,
940: the $(\theta,\phi)$-dependent part appears only in the second term,
941: so that the phase-difference angles are determined
942: as the minimum point of eq.~(\ref{ffie});
943: $(\theta,\varphi)=(0,\pi/2)$ for $g_{ff} <0$,
944: and $(\theta,\varphi)=(\pi,\pi/2)$ for $g_{ff} >0$.
945: It should be noted that the results are consistent with the condition
946: $\cos(\varphi)=\cos(\theta+\varphi)$
947: obtained in previous subsection.
948:
949:
950: \section{Classification of the density-wave structures in the ground state}
951:
952: Now, we analyze the minimum of the mean-field energy (\ref{ED3})
953: with the phase difference conditions obtained in the last part of sec.~II,
954: to determine the stable state of the system,
955: and show the complete classification of the ground state
956: for the dimensionless coupling constants $\zeta$ and $\xi$.
957:
958: \subsection{Repulsive fermion-fermion interaction $g_{ff}>0$}
959:
960: In the case of $g_{ff} >0$,
961: using $(\theta, \varphi)=(\pi, \pi/2)$
962: for the phase difference angles
963: in the ground state in eq.~(\ref{delbp0}),
964: we obtain
965: $\Delta_b'=-\frac{g_{ff}}{2 G_{bf}}\Delta_b \lk=-\frac{\xi}{2\zeta}\Delta_b\rk$.
966: Substituting them into eq.~(\ref{ED3}),
967: we obtain the total energy density up to the order of $O(\Delta^2)$:
968: %
969: \begin{eqnarray}
970: \tilde{\mathcal{E}}_{tot} = 2 n_f \epsilon_F\ltk -\frac{1}{2}
971: \left[ 1 -\log{\left(\frac{\tilde{\Delta}_B^2+\tdf^2}{4}\right)} \right]
972: \lk \tilde{\Delta}_B^2+\tdf^2 \rk
973: +\frac{2}{2\zeta-\xi}\tilde{\Delta}_B^2
974: +\frac{2}{\xi} \tdf^2 \rtk,
975: \label{eED1}
976: \end{eqnarray}
977: %
978: where $\tilde{\Delta}_B \equiv \lk 1-\frac{\xi}{2\zeta} \rk \Delta_b/(2\epsilon_F)$,
979: and $\tdf \equiv \Delta_f/(2\epsilon_F)$.
980:
981: In Fig.~\ref{Figa},
982: the contour plots of the energy density (\ref{eED1})
983: are shown in $\tilde{\Delta}_B$-$\tilde{\Delta}_f$ plane
984: for typical values of the coupling constants $\zeta$ and $\xi$.
985:
986: %------------------------------------------------------
987: \begin{figure}
988: \begin{center}
989: \includegraphics[height=13.5cm, angle=90]{spcpREPLi.eps}
990: \end{center}
991: \caption{Contour plots of $\tilde{\mathcal{E}}_{tot}$
992: in $\tilde{\Delta}_B$-$\tilde{\Delta}_f$ plane
993: for the repulsive Fermion-Fermion interaction $g_{ff}>0$.
994: Three plots are for $\zeta>\xi$ (a),
995: $\zeta=\xi$ (b)
996: and $\zeta<\xi$ (c). }
997: \label{Figa}
998: \end{figure}
999: %------------------------------------------------------
1000:
1001: Since the first term in eq.~(\ref{eED1}) is symmetric
1002: with respect to the order parameters $\tilde{\Delta}_{B,f}$,
1003: the position of the minimum point of eq.~(\ref{eED1})
1004: is determined by the last two terms;
1005: the results are summarized as
1006:
1007: \begin{itemize}
1008: %
1009: \item{$0<\xi<\zeta$ \qquad (Fig.~\ref{Figa}a)}, \qquad
1010: %
1011: $\lk \tdb,\tdf \rk
1012: = \lk 2\zeta \frac{2}{2\zeta-\xi}e^{\frac{-2}{2\zeta-\xi}}, 0 \rk$.
1013:
1014: In this case,
1015: the up- and down-fermions form in-phase density wave:
1016: %
1017: \begin{equation}
1018: \rho_\downarrow(z) = \rho_\uparrow(z)
1019: =n_f \ldk 1- \frac{\Delta_b}{2 \epsilon_F \zeta} \cos(2k_F z) \rdk,
1020: \label{fden1}
1021: \end{equation}
1022: %
1023: and the bosons make a periodic BEC:
1024: %
1025: \begin{equation}
1026: n_b(z) \simeq n_b \ldk 1 +\frac{\Delta_b}{ g_{bf} n_b} \cos(2k_F z) \rdk,
1027: \label{bden1}
1028: \end{equation}
1029: %
1030: which is analogous to charge-density waves (CDW) of electrons
1031: in low dimensional conductors \cite{kagoshima, CDW2}
1032: due to the Peierls instability.
1033: We here mention the roles which the sign of $g_{bf}$ plays in the density waves.
1034: Since the amplitude of the periodic BEC in (\ref{bden1}) is always positive
1035: by definition,
1036: the phase difference between boson and fermion density waves
1037: is determined by the sign of the amplitude of the fermion density wave
1038: in (\ref{fden1}),
1039: which is proportional to the sign of $g_{bf}$:
1040: attractive (repulsive) boson-fermion interaction
1041: for the in-phase (out of phase).
1042: %
1043: \item{$\zeta=\xi$ \qquad (Fig.~\ref{Figa}b)}, \qquad
1044: $\tilde{\Delta}_b^2/4+\tilde{\Delta}_f^2=4e^{\frac{-4}{\xi}}$.
1045:
1046: The above equation shows the existence of the rotation-like ``hidden'' symmetry
1047: in the ground state,
1048: which mixes between the bosonic and fermionic particle-hole pair order parameters.
1049: It also suggests the occurrence of the spontaneous symmetry breaking
1050: in the ground state
1051: and the existence of the Nambu-Goldstone mode
1052: as a low-lying excitation mode.
1053: %
1054: \item{$\zeta<\xi<2\zeta$ \qquad (Fig.~\ref{Figa}c) \ and\ $\xi \leq 2\zeta$}, \qquad
1055: $\lk \tdb,\tdf \rk = \lk 0, 2e^{\frac{-2}{\xi}} \rk$.
1056:
1057: In this case,
1058: the fermion-fermion interaction induces the out-phase fermion-density waves:
1059: %
1060: \begin{equation}
1061: \rho_{\uparrow, \downarrow}(z)
1062: =n_f\ldk 1 \pm \frac{\Delta_f}{\epsilon_F \xi} \sin(2k_F z) \rdk,
1063: \end{equation}
1064: %
1065: which forms the out-phase density wave
1066: in the two-component fermion system, analogous to the SDW \cite{ove, sdw1}.
1067:
1068: When $2\zeta\leq\xi$,
1069: we need a special care to treat the boson density wave,
1070: because the coefficient of $\tdb^2$ term in (\ref{eED1}) becomes negative,
1071: which can be traced back to the potential energy of the gap functions.
1072: Assembling the boson-fermion and fermion-fermion interaction terms (\ref{bfie}, \ref{ffie})
1073: and the boson energy (\ref{bmfe}),
1074: the potential energy up to the constant, $H_{pot}$, becomes
1075: %
1076: \begin{eqnarray}
1077: H_{pot}= -\frac{1}{G_{bf}} \lk 1
1078: -\frac{g_{ff}}{2G_{bf}} \rk \Delta_b^2
1079: -\frac{2}{g_{ff}}\Delta_f^2,
1080: \end{eqnarray}
1081: %
1082: which should be negative definite for any value of the gap functions
1083: to gain the condensation energy.
1084: Thus, it gives $\Delta_b \equiv 0$ for $2G_{bf}\le g_{ff}$ (or equivalently for $2\zeta\le \xi$).
1085:
1086: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1087:
1088:
1089: \end{itemize}
1090:
1091: \subsection{Attractive fermion-fermion interaction $g_{ff}<0$}
1092:
1093: In the case of $g_{ff}<0$,
1094: using the phase difference angles $(\theta, \varphi)=(0, \pi/2)$
1095: the relation $\tdb'=-\frac{\xi}{2\zeta}\tdb$ is obtained
1096: from eq.~(\ref{delbp0}) in the ground state.
1097: The total energy density (\ref{ED3}) becomes
1098: %
1099: \begin{eqnarray}
1100: \tilde{\mathcal{E}}_{tot} =2 n_f \epsilon_F
1101: \ltk -\frac{1}{2} \left[ 1 -\log{\left( \frac{\tilde{\Delta}_B^2+\tdf^2}{4} \right)}
1102: \right]
1103: \lk \tilde{\Delta}_B^2
1104: +\tdf^2 \rk
1105: +\frac{2}{2\zeta+|\xi|} \tilde{\Delta}_B^2
1106: +\frac{2}{|\xi|} \tdf^2 \rtk.
1107: \label{eED2}
1108: \end{eqnarray}
1109: %
1110: Fig.~\ref{Figc} shows a contour plot of the energy density
1111: in $\tilde{\Delta}_B$-$\tdf$ plane
1112: when $\zeta=0.5$ and $\xi=-0.6$.
1113:
1114: %------------------------------------------------------
1115: \begin{figure}
1116: \begin{center}
1117: \includegraphics[height=4.5cm]{spcpATT.eps}
1118: \end{center}
1119: \caption{Contour plots of $\tilde{\mathcal{E}}_{tot}$
1120: in $\tilde{\Delta}_B$-$\tilde{\Delta}_f$ plane
1121: for the attractive fermion-fermion interaction $g_{ff}<0$.}
1122: \label{Figc}
1123: \end{figure}
1124: %------------------------------------------------------
1125:
1126: As is clear from eq.~(\ref{eED2}),
1127: the coefficient of $\tilde{\Delta}_B^2$ is larger than that of $\tdf^2$,
1128: $2/(2\zeta+|\xi|) > 2/|\xi|$,
1129: so that the minimum point of the energy density is always
1130: on the $\tilde{\Delta}_B$ axis.
1131:
1132: Thus, for $g_{ff} <0$,
1133: the order parameters take values
1134: $\lk \tdb,\tdf \rk
1135: = \lk 2\zeta\frac{2}{2\zeta+|\xi|}e^{\frac{-2}{2\zeta+|\xi|}},0 \rk $,
1136: and fermion densities become in-phase,
1137: $\rho_\uparrow(z) = \rho_\downarrow(z)
1138: =n_f\ldk 1- \frac{\Delta_b}{2 \epsilon_F \zeta}
1139: \cos(2k_F z) \rdk$,
1140: which have the same form with those in eq.~(\ref{fden1})
1141: in the repulsive fermion-fermion case $g_{ff} >0$.
1142: The boson condensate has also the same form with (\ref{bden1}).
1143: However, the $|\xi|$-dependence of the order parameter $\Delta_b$
1144: is different in both cases as shown in Fig.~\ref{op1}
1145: where the gap functions are plotted for the effective coupling constant $\xi$
1146: when $\zeta=1$.
1147:
1148: %------------------------------------------------------
1149: \begin{figure}
1150: \begin{center}
1151: \includegraphics[height=6cm]{delta1.eps}
1152: \end{center}
1153: \caption{The gap functions (extreme values) at $\zeta=1$ as a function of $\xi$. }
1154: \label{op1}
1155: \end{figure}
1156: %------------------------------------------------------
1157:
1158: \subsection{Phase diagram}
1159:
1160: Finally, in Fig.~\ref{Figb},
1161: the phase diagram of the density-wave structure in the ground state
1162: is shown in the plane of the effective coupling constants $\zeta$ and $\xi$
1163: in eqs.~(\ref{vzeta},\ref{vxi}),
1164: which correspond to the $g_{bf}^2/g_{bb}$ and $g_{ff}$.
1165:
1166: In Fig.~\ref{Figb},
1167: there exist two phases denoted by in-phase and out-phase,
1168: which correspond to the fermionic in-phase density wave ($\theta=0$)
1169: and out-phase density wave ($\theta=\pi$)
1170: with/without the periodic BEC in the ground state.
1171:
1172: In the case of the repulsive fermion-fermion interaction ($\xi>0$),
1173: there exists a competition between in- and out-phase density waves,
1174: and the border of these two phases is given by $\xi=\zeta$
1175: (rigid line in Fig.~\ref{Figb}).
1176: Another critical border line $\xi=2\zeta$ exists
1177: (dotted line in Fig.~\ref{Figb}), and,
1178: above this line, the ground state has the out-phase fermion density waves
1179: but no periodic BEC.
1180:
1181: As for the attractive fermion-fermion interaction ($\xi<0$),
1182: the ground state has the periodic BEC
1183: which is coherent with the in-phase fermion density waves.
1184: The fermion-fermion interaction affects the boson magnitude $\Delta_b$ implicitly.
1185:
1186: The behaviors of the order parameters and the effective gap functions (\ref{egf1}-\ref{egf2})
1187: in Fig.~\ref{op1} are reflected
1188: in the phase diagram;
1189: $\Delta_f$ and $\Delta_b$ are continuous and smooth functions of $\xi$
1190: as stationary points of the energy density,
1191: and disappear at $\xi=0$ and $\xi=2\zeta$ respectively, as mentioned above.
1192: The line for the effective gap function $\Delta_{\uparrow, \downarrow}$
1193: (solid line),
1194: the net energy gap at the Fermi surfaces,
1195: bends at $\xi=\zeta$;
1196: it implies the transition from the in-phase density wave to the out-phase one.
1197:
1198: %------------------------------------------------------
1199: \begin{figure}
1200: \begin{center}
1201: \includegraphics[height=7cm]{SpinPeielPD2.eps}
1202: \end{center}
1203: \caption{The phase diagram of the density-wave structure
1204: of the ground state in $\zeta$-$\xi$ plane.}
1205: \label{Figb}
1206: \end{figure}
1207: %------------------------------------------------------
1208: %
1209:
1210: \section{Summary and Discussion}
1211:
1212: We studied the ground state of the Q1D boson-fermion mixture
1213: of the atomic gas with two-component fermion and one-component boson,
1214: and showed in the consistently formulated mean-field approximation
1215: that the states with the periodic BEC and
1216: the fermion density waves with the wave number $2k_F$
1217: become stable therein at $T=0$
1218: We also showed that
1219: there exists competition
1220: between the in-phase and out-phase fermion density waves
1221: in the ground state,
1222: which is originated from the relative strength
1223: between the boson-fermion and fermion-fermion effective interactions.
1224: We clarified the wave-density structures of the ground state
1225: and gave the phase diagram of it.
1226:
1227: Finally,
1228: we make a rough estimation for the density-wave amplitudes
1229: of the atomic gas mixtures in the axially deformed trap potential.
1230: As possible candidates,
1231: we take rubidium isotope mixtures,
1232: ${}^{84}{\rm Rb}$-${}^{87}{\rm Rb}$ and ${}^{86}{\rm Rb}$-${}^{87}{\rm Rb}$,
1233: with the parameters $\omega_a=2 \pi \times 10~{\rm Hz}$ and
1234: $\omega_r=2 \pi \times 15~{\rm kHz}$ for the trap potential,
1235: and $N_f\simeq 10^3$ and $N_b\simeq 2 \times 10^4$ for the atom numbers,
1236: with which the Peierls instability is shown to exist in \cite{MYS1}.
1237: For the scattering lengths in atomic collisions,
1238: we take the values in \cite{est1} (atomic unit):
1239: %
1240: \begin{eqnarray}
1241: &&\mbox{1)~${}^{84}{\rm Rb}$-${}^{87}{\rm Rb}$:}\quad
1242: (a_{ff}, a_{bf}, a_{bb})=(142, 117, 90)
1243: \Rightarrow (\zeta, \xi)=(0.397, 0.467), \\
1244: %
1245: &&\mbox{2)~${}^{86}{\rm Rb}$-${}^{87}{\rm Rb}$:}\quad
1246: (a_{ff}, a_{bf}, a_{bb})=(7, 336, 90)
1247: \Rightarrow (\zeta, \xi)=(3.276, 0.023).
1248: \end{eqnarray}
1249: %
1250: Using these parameters,
1251: we obtain the results that,
1252: in ${}^{84}{\rm Rb}$-${}^{87}{\rm Rb}$ mixture (the case 1)),
1253: the out-phase density wave should occur with the amplitude
1254: $\Delta_f/(\epsilon_F \xi) \simeq 0.118$,
1255: and, in ${}^{86}{\rm Rb}$-${}^{87}{\rm Rb}$ mixture (the case 2)),
1256: the ground state has the in-phase one with the amplitude
1257: $\Delta_f/(2\epsilon_F \zeta) \simeq 0.451$
1258: and periodic BEC with amplitudes $\Delta_b/(g_{bf} n_b) \simeq 0.321$.
1259: The changes of the atom number, the trap potential size
1260: and also the interaction strengths should make the shift of
1261: these results, and make the phase transitions shown in Fig.~6
1262: expected to be observable.
1263:
1264: %For experimental verification of the phase transition
1265: %the observation of the collective excitation modes
1266: %are very important,
1267: %because they should show the instability at the transition
1268: %
1269: %In future works,
1270: %it is interesting to investigate collective modes of the system
1271: %in the normal phase and in the density wave phases.
1272: %The former should show instability modes corresponding to the boson-fermion %density waves of the present study,
1273: %and the latter gives dynamical properties of the density wave phases %\cite{CDW2, sdw1, LRA}.
1274: %Furthermore,
1275: %we are interested in periodically trapped systems of the mixture
1276: %in which the periodic trap potential gives band structure analogous to metals,
1277: %and locking of the density waves to the external potential is expected %\cite{kagoshima}.
1278:
1279: \begin{appendix}
1280:
1281: \section{Diagonalization of the mean-field Hamiltonian}
1282: In this appendix,
1283: we briefly sketch the diagonalization
1284: of the mean-field Hamiltonian (\ref{MFH1})
1285: under the periodic BEC and the fermion density waves
1286: given in eqs.~(\ref{BD1}) and (\ref{FD2}),
1287: and show the quasi-particle wave functions.
1288:
1289: The up- and down-fermion contributions in eq.~(\ref{MFH1})
1290: have the same form in structure
1291: and the calculations go completely in parallel,
1292: so that we omit the subscript of the fermion field operator
1293: in this appendix.
1294: The first and second terms in eq.~(\ref{MFH1}) should be
1295: diagonalized by a unitary matrix $U$:
1296: %
1297: \begin{eqnarray}
1298: && \left(\begin{array}{c}
1299: \psi(k) \\
1300: \psi(k+2k_F)
1301: \end{array}\right)^\dagger
1302: U U^{-1}
1303: \left[ \begin{array}{c c}
1304: \epsilon(k) & \Delta \\
1305: \Delta^* & \epsilon(k+2k_F)
1306: \end{array} \right]
1307: U U^{-1}
1308: \left(\begin{array}{c}
1309: \psi(k) \\
1310: \psi(k+2k_F)
1311: \end{array}\right) \nn
1312: %
1313: &&\qquad\qquad
1314: = \left(\begin{array}{c}
1315: {\alpha^-(k)} \\
1316: {\alpha^+(k)}
1317: \end{array}\right)^\dagger
1318: \left[\begin{array}{c c}
1319: E^-(k) & 0 \\
1320: 0 & E^+(k)
1321: \end{array}\right]
1322: \left(\begin{array}{c}
1323: \alpha^-(k) \\
1324: \alpha^+(k)
1325: \end{array}\right),
1326: \end{eqnarray}
1327: %
1328: where $E^\pm(k)$ are the quasi-particle spectrum
1329: given in eqs.~(\ref{qpe1}) and (\ref{qpe2}),
1330: and $\alpha^\pm(k)$ is the field operator
1331: corresponding to them.
1332:
1333: The explicit form of the unitary matrix is given by
1334: %
1335: \begin{equation}
1336: U=\left[ \begin{array}{c c}
1337: \tilde{v}(k) & \tilde{u}(k) \\
1338: u(k) & v(k)
1339: \end{array}\right], \qquad
1340: %
1341: U^{-1}=U^\dagger=\left[ \begin{array}{c c}
1342: \tilde{v}^*(k) & u^*(k) \\
1343: \tilde{u}^*(k) & v^*(k)
1344: \end{array}\right],
1345: \end{equation}
1346: %
1347: where
1348: %
1349: \begin{eqnarray}
1350: && \tilde{u}(k) =\frac{\Delta}{|\Delta|} \sqrt{\frac{1}{2}}
1351: \ldk 1 +\frac{\epsilon(k)
1352: -\epsilon(k+2k_F)}{ \sqrt{\ltk \epsilon(k)
1353: -\epsilon(k+2k_F) \rtk^2
1354: +|\Delta|^2}} \rdk^{1/2} \\
1355: %
1356: && \tilde{v}(k) =-\frac{\Delta}{|\Delta|} \sqrt{\frac{1}{2}}
1357: \ldk 1 -\frac{\epsilon(k)
1358: -\epsilon(k+2k_F)}{\sqrt{\ltk \epsilon(k)
1359: -\epsilon(k+2k_F) \rtk^2
1360: +|\Delta|^2}} \rdk^{1/2} \\
1361: %
1362: && u(k) =\sqrt{\frac{1}{2}}
1363: \ldk 1 +\frac{\epsilon(k)
1364: -\epsilon(k+2k_F)}{\sqrt{\ltk \epsilon(k)
1365: -\epsilon(k+2k_F) \rtk^2
1366: +|\Delta|^2}} \rdk^{1/2} \\
1367: %
1368: && v(k) =\sqrt{\frac{1}{2}}
1369: \ldk 1-\frac{\epsilon(k)
1370: -\epsilon(k+2k_F)}{\sqrt{\ltk \epsilon(k)
1371: -\epsilon(k+2k_F)\rtk^2
1372: +|\Delta|^2}} \rdk^{1/2},
1373: \end{eqnarray}
1374: %
1375: for $-k_F\le k \le 0$.
1376: The quasi-particle wave functions, $u(k)$ and $v(k)$,
1377: have been normalized to satisfy the condition:
1378: $|u(k)|^2+|v(k)|^2=|\tilde{u}(k)|^2+|\tilde{v}(k)|^2=1$.
1379:
1380: \section{Gap equations}
1381:
1382: The gap equations of the system are given by
1383: %
1384: \begin{eqnarray}
1385: &&-\frac{4}{g_{ff}} \ltk \Delta_f \cos(\theta)+
1386: \Delta_b' \cos\left(\frac{\theta}{2} \right)
1387: \cos\left(\varphi+\frac{\theta}{2} \right)
1388: \rtk \nn
1389: %
1390: &&\qquad= 2 \int_{-k_F}^0 \frac{dk}{2\pi}
1391: \ltk \frac{(\Delta_b+\Delta_b') \cos(\varphi)
1392: +\Delta_f}{\sqrt{ \ltk \epsilon(k)
1393: -\epsilon(k+2k_F)\rtk^2
1394: +|\Delta_\uparrow|^2 }}
1395: +\frac{(\Delta_b+\Delta_b') \cos(\varphi+\theta)
1396: +\Delta_f}{\sqrt{ \ltk \epsilon(k)
1397: -\epsilon(k+2k_F) \rtk^2
1398: +|\Delta_\downarrow|^2 }} \rtk \\
1399: %
1400: {}\nn
1401: %
1402: &&\frac{2\Delta_b}{G_{bf}}
1403: %
1404: = 2 \int_{-k_F}^0 \frac{dk}{2\pi}
1405: \ltk \frac{\Delta_f \cos(\varphi)
1406: +\Delta_b
1407: +\Delta_b'}{\sqrt{ \ltk \epsilon(k)
1408: -\epsilon(k+2k_F)\rtk^2
1409: +|\Delta_\uparrow|^2 }}
1410: +\frac{\Delta_f \cos(\varphi+\theta)
1411: +\Delta_b
1412: +\Delta_b'}{\sqrt{ \ltk \epsilon(k)
1413: -\epsilon(k+2k_F) \rtk^2
1414: +|\Delta_\downarrow|^2 }} \rtk \\
1415: %
1416: {}\nn
1417: %
1418: &&-\frac{4}{g_{ff}} \ltk \Delta_b'
1419: +\Delta_f \cos\left(\frac{\theta}{2} \right)
1420: \cos\left(\varphi+\frac{\theta}{2} \right)
1421: \rtk \nn
1422: %
1423: &&\qquad= 2 \int_{-k_F}^0 \frac{dk}{2\pi}
1424: \ltk \frac{\Delta_f \cos(\varphi)
1425: +\Delta_b
1426: +\Delta_b'}{\sqrt{ \ltk \epsilon(k)
1427: -\epsilon(k+2k_F)\rtk^2
1428: +|\Delta_\uparrow|^2 }}
1429: +\frac{\Delta_f \cos(\varphi+\theta)
1430: +\Delta_b
1431: +\Delta_b'}{\sqrt{ \ltk \epsilon(k)
1432: -\epsilon(k+2k_F) \rtk^2
1433: +|\Delta_\downarrow|^2 }}\rtk
1434: \end{eqnarray}
1435: %
1436: which can be derived from the energy density eq.~(\ref{ED1})
1437: by taking the variation with respect to the gap functions
1438: $\Delta_f$, $\Delta_b$, and $\Delta_b'$ each other.
1439:
1440: \end{appendix}
1441:
1442: %
1443: \begin{thebibliography}{99}
1444: \bibitem{BECexp1}
1445: M. H. Anderson, J. R. Ensher, M. R. Matthews, C. E. Wieman,
1446: and E. A. Cornell, Science 269, 198 (1995); K. B. Davis,
1447: M.-O. Mewes, M. R. Andrews, N. J. van Druten, D. S. Durfee,
1448: D. D. Kurn, and W. Ketterle, Phys. Rev. Lett. 75, 3969 (1995).
1449: %
1450: \bibitem{BECtext1}
1451: F. Dalfovo, S. Giorgini, L. P. Pitaevskii, and S. Stringari,
1452: Rev. Mod. Phys. {\bf 71}, 463 (1999)
1453: %
1454: \bibitem{FDexp1}
1455: B. DeMarco and D. S. Jin, Science {\bf 285}, 1703 (1999);
1456: B. DeMarco, S. B. Rapp, and D. S. Jin, Phys. Rev. Lett. {\bf 86}, 5409 (2001).
1457: %
1458: \bibitem{BFMexp2}
1459: A. G. Truscott, K. E. Strecker, W. I. McAlexander, G. B. Partridge, and R. G. Hulet,
1460: Science {\bf 291} 2570 (2001).
1461: %
1462: \bibitem{BFMexp1}
1463: F. Schreck, L. Khaykovich, K. L. Corwin, G. Ferrari, T. Bourdel, J. Cubizolles, and C. Salomon,
1464: Phys. Rev. Lett. {\bf 87} 080403 (2001).
1465: %
1466: \bibitem{BFMtheory1}
1467: K. M\o lmer, Phys. Rev. Lett. {\bf 80}, 1804 (1998).
1468: %
1469: \bibitem{BFMtheory2}
1470: T. Tsurumi and M. Wadati, J. Phys. Soc. Jpn. {\bf 69}, 97 (2000).
1471: %
1472: \bibitem{BFMtheory3}
1473: A. Minguzzi and M. P. Tosi, Phys. Lett. A {\bf 268}, 142 (2000).
1474: %
1475: \bibitem{BFMtheory4}
1476: M. J. Bijlsma, B. A. Heringa, and H. T. C. Stoof, Phys. Rev. A {\bf 61}, 053601 (2000).
1477: %
1478: \bibitem{BFMtheory5}
1479: L. Viverit, C. J. Pethick, and H. Smith, Phys. Rev. A {\bf 61}, 053605 (2003).
1480: %
1481: \bibitem{BFMtheory6}
1482: R. Roth and H. Feldmeier, Phys. Rev. A {\bf 65}, 021603(R) (2002).
1483: %
1484: \bibitem{BFMtheory7}
1485: C. P. Search, H. Pu, W. Zhang, and P. Meystre, Phys. Rev. A {\bf 65}, 063615 (2002).
1486: %
1487: \bibitem{BFMtheory8}
1488: T. Sogo, T. Miyakawa, T. Suzuki, and H. Yabu, Phys. Rev. A {\bf 66}, 013618 (2002).
1489: %
1490: \bibitem{BFMtheory9}
1491: H. P. B\"{u}chler and G. Blatter, Phys. Rev. Lett. {\bf 91}, 130404 (2003).
1492: %
1493: \bibitem{MYS1}
1494: T. Miyakawa, H. Yabu, and T. Suzuki,
1495: Phys. Rev. A {\bf 70} 013612 (2004); Physica {\bf B 329-333}, 28 (2003).
1496: %
1497: \bibitem{tonk1}
1498: L. Tonks, Phys. Rev. {\bf 50}, 955 (1936).
1499: %
1500: \bibitem{tonk2}
1501: M. Girardeau, J. Math. Phys. (N.Y.) {\bf 1}, 516 (1960).
1502: %
1503: \bibitem{regime1}
1504: D. S. Petrov, G.V. Shlyapnikov, and J. T. M. Walraven,
1505: Phys. Rev. Lett. {\bf 85} 3745 (2000).
1506: %
1507: \bibitem{regime2}
1508: I. V. Tokatly, cond-mat/0402276.
1509: %
1510: \bibitem{Das1}
1511: K. K. Das,
1512: Phys. Rev. A {\bf 66} 053612 (2002);
1513: Phys. Rev. Lett. {\bf 90} 170403 (2003).
1514: %
1515: \bibitem{Pita1}
1516: L. Pitaevskii and S. Stringari,
1517: {\it Bose-Einstein Condensation}
1518: (Oxford University Press, London, 2003)
1519: %
1520: \bibitem{BEClow1}
1521: A. G\"orlitz, J. M. Vogels, A. E. Leanhardt, C. Raman, T. L. Gustavson, J. R. Abo-Shaeer,
1522: A. P. Chikkatur, S. Gupta, S. Inouye, T. Rosenband, and W. Ketterle,
1523: Phys. Rev. Lett. {\bf 87}, 160405 (2001)
1524: %
1525: \bibitem{BEClow2}
1526: M. Greiner, I. Bloch, O. Mandel, T. W. H\"ansch, and T. Esslinger,
1527: Phys. Rev. Lett. {\bf 87}, 130402 (2001)
1528: %
1529: \bibitem{BEClow3}
1530: H. Moritz, T. Stoferle, M. Kohl, and T. Esslinger,
1531: Phys. Rev. Lett. {\bf 91}, 250402 (2003)
1532: %
1533: \bibitem{soliton1}
1534: J. Denschlag, J. E. Simsarian, D. L. Feder, Charles W. Clark,
1535: L. A. Collins, J. Cubizolles, L. Deng, E. W. Hagley,
1536: K. Helmerson, W. P. Reinhardt, S. L. Rolston, B. I. Schneider,
1537: W. D. Phillips,
1538: Science {\bf 287}, 97 (2000).
1539: %
1540: \bibitem{soliton2}
1541: L. Khaykovich, F. Schreck, G. Ferrari, T. Bourdel,
1542: J. Cubizolles, L. D. Carr, Y. Castin, C. Salomon,
1543: Science {\bf 296}, 1290 (2002).
1544: %
1545: \bibitem{soliton3}
1546: J. Ieda, T. Miyakawa, and M. Wadati,
1547: Phys. Rev. Lett. {\bf 93}, 194102 (2004).
1548: %
1549: \bibitem{peierl}
1550: R. E. Peierls, {\it Quantum Theory of Solids}
1551: (Oxford University Press, London, 1955).
1552: %
1553: \bibitem{Kittel}
1554: C. Kittel, {\it Quantum Theory of Solids}
1555: (John Wiley \& Sons Inc, New York, 1996).
1556: %
1557: \bibitem{kagoshima}
1558: S. Kagoshima, H. Nagasawa, and T. Sambongi,
1559: {\it One Dimensional Conductors},
1560: Springer series in solid-state sciences, Vol. 72
1561: (Springer-Verlag, Berlin, 1988);
1562: L. P. Gor'kov and G. Gr\"uner, {\it Charge Density waves in Solids},
1563: MODERN PROBLEMS IN CONDENSED MATTER SCIENCES. VOL. 25
1564: (AMSTERDAM: North-Holland, 1989).
1565: %
1566: \bibitem{pethick}
1567: For example, C. J. Pethick and H. Smith,
1568: {\it Bose-Einstein Condensation in Dilute Gases}
1569: (Cambridge University Press, UK, 2002)
1570: %
1571: \bibitem{SpinPol}
1572: T. Sogo and H. Yabu,
1573: Phys. Rev. A {\bf 66}, 043611 (2002).
1574: %
1575: \bibitem{CDW2}
1576: G. Gr\"{u}ner, Rev. Mod. Phys. {\bf 60}, 4 (1988).
1577: %
1578: \bibitem{ove}
1579: A.W. Overhauser,
1580: Phys. Rev. Lett.{\bf 4}, 462 (1960);
1581: Phys. Rev. {\bf 128}, 1437 (1962).
1582: %
1583: \bibitem{sdw1}
1584: G. Gr\"{u}ner, Rev. Mod. Phys. {\bf 66}, 1 (1994).
1585: %
1586: \bibitem{est1}
1587: J. P. Burke, Jr. and J. L. Bohn, Phys. Rev.{\bf A 59}, 1303 (1999).
1588: %
1589: %\bibitem{LRA}
1590: %P. A. Lee, T. M. Rice, and P. M. Anderson,
1591: %Phys. Rev. lett. {\bf 31}, 462 (1973).
1592: \end{thebibliography}
1593: %
1594:
1595: \end{document}
1596: