1:
2:
3: \documentclass[12pt,a4paper]{iopart}
4: \usepackage{amssymb,graphicx,cite}
5:
6:
7: %LATEX FILE OF MANUSCRIPT
8: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
9:
10:
11:
12: %LATEX file of the manuscript
13:
14: %\documentclass[preprint,eqsecnum,aps,epsf]{revtex4} % PH. REV.
15: %\documentclass[eqsecnum,aps,twocolumn,epsf,showpacs]{revtex4} % PH.
16: %REV.
17: %\documentclass[aps,twocolumn,epsf,showpacs,floatfix]{revtex4} % PH. REV.
18: \usepackage{graphicx}
19: %\documentclass[eqsecnum,aps,twocolumn]{revtex4} % PH. REV.
20:
21: %\renewcommand{\baselinestretch}{.89}
22:
23:
24: %\documentstyle[aps,epsf]{revtex}
25: %\documentstyle[eqsecnum,aps,epsf]{revtex}
26: %%% <<< epsf commands in the next two lines >>>
27: %\newcommand{\postscript}[2] {\setlength{\epsfxsize}{#2\hsize}
28: %\centerline{\epsfbox{#1}}}
29:
30:
31:
32: %\documentstyle[eqsecnum,aps,epsf]{revtex} % PH. REV. FINAL FORMAT STYLE
33: %\documentstyle[aps,epsf]{revtex} % PH. REV. FINAL FORMAT STYLE
34: %%% <<< epsf commands in the next two lines >>>
35: %\newcommand{\postscript}[2] {\setlength{\epsfxsize}{#2\hsize}
36: %\centerline{\epsfbox{#1}}}
37:
38: %\documentstyle[preprint,aps]{revtex}
39: %\documentstyle[eqsecnum,aps]{revtex}
40: %\documentstyle[aps]{revtex}
41: %\renewcommand{\baselinestretch}{.945}
42:
43: \begin{document}
44:
45: %\twocolumn[\hsize\textwidth\columnwidth\hsize\csname@twocolumnfalse\endcsname
46:
47: %\title[Jet formation in a collapsing Bose-Einstein condensate]{Mean-field
48: %model of jet formation in a collapsing Bose-Einstein condensate}
49:
50:
51:
52:
53: \title[Bright solitons in a
54: fermion-fermion mixture]{Formation of bright solitons and soliton
55: trains in a
56: fermion-fermion mixture by modulational instability}
57:
58: \author{Sadhan K. Adhikari\footnote{Electronic
59: address: adhikari@ift.unesp.br; \\
60: URL: http://www.ift.unesp.br/users/adhikari/}}
61: %\affiliation
62: \address
63: {Instituto de F\'{\i}sica Te\'orica, UNESP $-$ S\~ao Paulo
64: State University, 01.405-900 S\~ao Paulo, S\~ao Paulo, Brazil}
65:
66: \date{\today}
67:
68: %\maketitle
69:
70: \begin{abstract}
71:
72: We employ a time-dependent mean-field-hydrodynamic
73: model to study the generation of bright solitons in a
74: degenerate
75: fermion-fermion mixture in a cigar-shaped geometry
76: using variational and numerical methods.
77: Due to a strong Pauli-blocking repulsion among identical spin-polarized
78: fermions at short distances there cannot be bright solitons for
79: repulsive interspecies interactions. Employing a linear stability
80: analysis
81: we demonstrate the formation of stable
82: solitons due to modulational instability of a
83: constant-amplitude solution
84: of the model equations
85: for a sufficiently
86: attractive interspecies interaction.
87: We perform a
88: numerical stability analysis of these solitons and also demonstrate the
89: formation of soliton trains by jumping the effective interspecies
90: interaction from repulsive to attractive.
91: These fermionic solitons can be formed and
92: studied in laboratory with present technology.
93:
94:
95:
96: \pacs{ 03.75.Ss, 05.45.Yv}
97:
98:
99:
100: \end{abstract}
101:
102: \maketitle
103:
104:
105:
106:
107:
108: %\keywords{Degenerate Fermi gas, Bright soliton, Soliton train}
109:
110:
111: %\date{\today}
112:
113:
114:
115:
116: %\maketitle
117:
118: \section{Introduction}
119:
120: After experimental observation and the study of bright solitons in
121: Bose-Einstein condensates (BEC) \cite{exdks1,exdks2,yuka},
122: recent observations \cite{exp1,exp2,exp3,exp4} and
123: experimental \cite{exp5,exp5x,exp6} and theoretical
124: \cite{yyy1,yyy,capu,ska} studies
125: of a degenerate Fermi gas (DFG) by
126: sympathetic cooling in
127: the presence of a second boson or fermion component suggest the
128: possibility of soliton formation \cite{vpg} in a degenerate
129: fermion-fermion mixture (DFFM).
130: Apart from the observation of a DFG in the
131: degenerate boson-fermion
132: mixtures (DBFM)
133: $^{6,7}$Li
134: \cite{exp3}, $^{23}$Na-$^6$Li \cite{exp4} and $^{87}$Rb-$^{40}$K
135: \cite{exp5,exp5x}, there have been studies of the following
136: spin-polarized
137: DFFM
138: $^{40}$K-$^{40}$K \cite{exp1} and $^6$Li-$^6$Li \cite{exp2}.
139:
140:
141:
142: The dimensionless
143: one-di\-mensional
144: nonlinear Schr\"odinger (NLS)
145: equation in the attractive (self-focussing) case \cite{1}
146: \begin{equation}\label{nls} i \frac{\partial \phi}{\partial
147: t}+\frac{1}{2}\frac {\partial^2 \phi}{\partial y^2}+ |\phi|^2\phi =0.
148: \end{equation} sustains the following bright
149: soliton \cite{1}:
150: \begin{eqnarray}\label{DS}
151: \phi(y,t)&=& a \hskip 3pt \mbox{sech}
152: [a(y-v t)] e^{ivy -i(v^2 -a^2)t/2+i\sigma}
153: \end{eqnarray}
154: The parameter $a$ represents the amplitude as well as pulse width, $v$
155: the velocity, and $\sigma$ is a phase constant. These bright solitons
156: are possible only in one dimension. In two and three dimensions they are
157: allowed in the presence of a transverse trap \cite{exdks1,vpg}, or
158: in the presence of
159: an
160: oscillating nonlinearity \cite{soldy}.
161: Apart from bright solitons in the attractive case, gap solitons are
162: possible in the repulsive (self-defocussing) case in the presence of a
163: periodic potential \cite{gs}.
164:
165:
166: Bright solitons in a
167: BEC are formed due to a nonlinear atomic
168: attraction \cite{exdks1,exdks2}. As the interaction in a pure DFG at
169: short
170: distances is
171: repulsive due to strong Pauli blocking, there cannot be bright solitons in
172: a DFG.
173: However, bright
174: solitons
175: can be
176: formed \cite{fbs1,fbs2}
177: in a DBFM in the presence of a sufficiently strong
178: boson-fermion attraction which can overcome the Pauli repulsion among
179: identical fermions. Bright solitons can also be formed in a binary
180: mixture of repulsive bosons supported by interspecies attraction
181: \cite{perez}.
182:
183:
184:
185: We demonstrate the formation of stable fermionic bright solitons in a DFFM
186: for a sufficiently attractive interspecies interaction.
187: In a DFFM, the coupled system can lower its energy by forming high density
188: regions, the bright solitons, when the attraction between the two types of
189: fermions is large enough to overcome the Pauli repulsion. We use a coupled time-dependent mean-field-hydrodynamic model
190: for a DFFM and consider the formation of axially-free localized bright
191: solitons in a quasi-one-dimensional cigar-shaped geometry using numerical
192: and variational solutions. The present model is inspired by the success
193: of a similar model used recently in the investigation of collapse
194: \cite{ska,skal} and bright \cite{fbs2} and dark \cite{fds} solitons in a
195: DBFM, and black solitons \cite{bla}
196: and mixing-demixing \cite{mix} in a DFFM.
197:
198:
199:
200:
201:
202:
203:
204:
205:
206:
207: We study the condition of modulational instability of a constant-amplitude
208: solution of the present model under a plane-wave perturbation and
209: demonstrate the possibility of the formation of bright solitons by a
210: linear stability analysis. We find that for a sufficiently strong
211: interspecies interaction, under the plane-wave perturbation the
212: constant-amplitude solution becomes unstable and localized solitonic
213: solutions may appear. We present a numerical stability analysis of these
214: bright solitons by introducing different small perturbations when the
215: solitons undergo stable and sustained breathing oscillation.
216:
217: We also
218: consider and study the formation of a soliton train in a DFFM by a large
219: sudden jump in the interspecies fermion-fermion scattering length
220: realized by manipulating a background magnetic field near a Feshbach
221: resonance, experimentally observed in two hyperfine
222: states of the
223: following spin-polarized
224: DFFMs: $^6$Li-$^6$Li and $^{40}$K-$^{40}$K \cite{fesh}.
225: This procedure effectively and rapidly turns a repulsive DFFM to a
226: highly attractive one and generates bright solitons.
227: An experimental realization of bright solitons in a DFFM could be tried
228: in two hyperfine states in samples such as $^6$Li-$^6$Li or
229: $^{40}$K-$^{40}$K using a Feshbach resonance.
230: However, there is already experimental evidence \cite{gehm} and
231: theoretical
232: conjecture \cite{kkk}
233: that the maximum attractive force, that can be created by a
234: Feshbach resonance in fermionic atoms in two hyperfine states is
235: limited
236: by quantum mechanical constraints of unitarity. Although there is a
237: limit to the creation of attraction in a two-component spin-polarized
238: DFFM in two hyperfine states, there is no such limit in a
239: multi-component
240: spin-polarized
241: DFFM \cite{hei}
242: or a DFFM composed of atoms of distinct mass \cite{hulet1}.
243: So, if bright solitons and soliton trains cannot be efficiently created
244: in a two-component DFFM in two hyperfine states,
245: a better and more efficient DFFM for
246: the creation of solitons could be the mixture of
247: fermionic atoms of distinct mass, where one can avoid the problem of a
248: possible suppression of interspecies attraction. One such system, among
249: many others, could
250: be the mixture of spin-polarized $^6$Li-$^{40}$K mixture: both $^6$Li
251: \cite{exp2,exp211}
252: and $^{40}$K \cite{exp1} have
253: been trapped and studied in laboratory.
254: The formation of bright solitons in a DFFM by turning the effective
255: interspecies interaction among spin-polarized
256: fermions from repulsion to strong attraction
257: seems within the reach of present experimental possibilities.
258:
259:
260: Here we shall be interested in the formation of bright solitons in a
261: spin-polarized DFFM of different fermionic atoms in the presence of
262: strong interspices attraction and we shall ignore the possibility of the
263: formation of a Bardeen-Cooper-Schreiffer (BCS) condensate \cite{bcs}
264: through Cooper pairing. A BCS condensate is usually formed in the
265: presence of a weak attraction among identical fermions with fermion
266: pairs in the singlet (spin parallel) state. The formation of a BCS
267: condensate should not be favored in a spin-polarized (spin antiparallel)
268: two-component DFFM with strong attraction among the different types of
269: atoms, as a strong interspecies attraction is not the domain of BCS
270: condensation. By choosing a strong interspecies attraction our study
271: stays in the BEC region of the BCS-Bose crossover problem \cite{BCSB}
272: strongly favoring molecule formation. The solitons once formed will
273: decay via molecule formation of two different types of fermionic atoms
274: \cite{hulet3} as in the case of bosonic solitons. Nevertheless, such
275: molecule formation is a slow process and can be accounted for by a
276: three-body recombination term as introduced in a previous study of
277: collapse in a DFFM \cite{njp}. The molecule formation will eventually
278: destroy the solitons in the DFFM, albeit, at a slow rate. The same is
279: true in the formation of soliton and soliton train in an attractive BEC,
280: where the essential features of the dynamics have been well explained
281: within the mean-field Gross-Pitaevskii model by neglecting molecule
282: formation \cite{hulet2}. A subsequent study including the effect of
283: molecule formation \cite{sala2} has not essentially changed the
284: conclusions of \cite{hulet2}. Hence this pioneering study on the
285: formation of soliton and soliton trains in a DFFM of different atoms
286: using mean-field hydrodynamic equations with the neglect of molecule
287: formation is expected to explain the essential features of the dynamics.
288: However, it would be of interest to include the effect of molecule
289: formation on the DFFM solitons in a future study.
290:
291:
292:
293:
294:
295:
296: In section 2 we present our mean-field-hydrodynamic model and its
297: reduction
298: to a quasi-one-dimensional form in a cigar-shaped geometry. In section
299: 3
300: we show that bright solitons can appear in this model
301: through modulational instability of a constant amplitude solution. In
302: section 4 we perform a variational analysis of the mean-field equations.
303: Numerical results for isolated bright solitons are presented in section
304: 5 and
305: are compared with variational results. Then we study the generation of
306: a
307: train of solitons by modulational instability. Finally, we present a
308: summary in section 6.
309:
310: \section{Nonlinear Model}
311:
312:
313: We use a simplified mean-field-hydrodynamic Lagrangian for
314: a DFG used successfully to study a DBFM
315: \cite{ska,fbs2,fds}.
316: %The virtue of the
317: %mean-field model over a microscopic description is its simplicity and
318: %predictive power.
319: To develop a set of time-dependent
320: mean-field-hydrodynamic
321: equations for the interacting DFFM, we use the
322: following Lagrangian density \cite{ska,fbs2}
323: \begin{eqnarray}\label{yy} {\cal
324: L}= g_{12}n_1n_2+\sum_{j=1}^2
325: \frac{i}{2}\hbar \left[ \psi_j\frac{\partial {\psi_j} ^*}{\partial
326: t} - {\psi_j}^* \frac{\partial \psi_j}{\partial t} \right]+
327: \end{eqnarray} \begin{eqnarray}
328: + \sum_{j=1}^2
329: \left(\frac{\hbar^2 |\nabla_{\bf r} \psi_j|^2 }{6m_j}+
330: V_j({\bf r})n_j+\frac{3}{5} A_j n_j^{5/3}\right),\nonumber
331: \end{eqnarray}
332: where $j=1,2$ represents the two components, $\psi_j$ the probability
333: amplitude, $n_j=|\psi_j| ^2$ the probability
334: density,
335: $^*$ denotes complex conjugate, $m_j$ the
336: mass,
337: $A_j=\hbar^2(6\pi^2)^{2/3}/(2m_i),$
338: the interspecies coupling
339: $g_{12}=2\pi \hbar^2 a_{12}
340: /m_R$
341: with $m_R=m_1m_2/(m_1+m_2)$ the reduced mass, and
342: $ a_{12}$
343: the interspecies
344: fermion-fermion scattering length.
345: The
346: number of fermionic atoms $N_j$
347: is given by $\int d{\bf r} n_j({\bf r})=N_j$.
348: The trap potential with axial symmetry is
349: taken as $
350: V_{j}({\bf
351: r})=\frac{1}{2} 3m_j \omega ^2 (\rho^2+\nu^2 z^2)$ where
352: $\omega$ and $\nu \omega$ are the angular frequencies in the radial
353: ($\rho$) and axial ($z$) directions with $\nu$ the anisotropy.
354: The $\nu \to 0$ limit corresponds to a cigar-shaped geometry and
355: allows a
356: reduction of the three-dimensional equations to a quasi-one-dimensional
357: form appropriate for freely moving solitons.
358: The interaction between identical intra-species fermions in
359: spin-polarized state is highly suppressed due
360: to Pauli blocking terms $3A_jn_j^{5/3}/5$
361: and has been neglected in (\ref{yy}). The
362: kinetic energy terms $\hbar^2|\nabla_{\bf r}\psi_j|^2$
363: $/(6m_j)$
364: in (\ref{yy})
365: contribute little to this problem compared to the
366: dominating Pauli-blocking terms.
367: However, its inclusion leads
368: to an analytic solution for the probability density everywhere
369: \cite{fbs2}.
370:
371:
372:
373: With the Lagrangian density (\ref{yy}), the following Euler-Lagrange
374: equations:
375: \begin{equation}
376: \frac{d}{d t}\frac{\partial {\cal L}}{\partial
377: \frac{\partial
378: \psi_j\- ^*}{\partial t}}+
379: \sum _{k=1}^3 \frac{d}{dx_k}\frac{\partial {\cal L}}{\partial
380: \frac{\partial \psi_j\- ^*}{\partial x_k}}= \frac{\partial {\cal
381: L}}{\partial
382: \psi_j\- ^*},
383: \end{equation}
384: with $x_k, k=1,2,3$ being the three space components,
385: become
386: \begin{eqnarray}\label{e} \biggr[
387: i\hbar\frac{\partial }{\partial t} +\frac{\hbar^2\nabla_{\bf
388: r}^2}{6m_{{j}}} - V_{{j}} - A_jn_j^{2/3}-
389: g_{{12}}
390: n_k
391: \biggr]\psi_j=0,
392: \end{eqnarray}
393: where $j\ne k = 1,2$. This is a time-dependent version of a
394: similar time-independent hydrodynamic
395: model for fermions \cite{capu}.
396: For large $n_j$, both lead to \cite{capu,ska}
397: the
398: Thomas-Fermi result \cite{yuka} $n_j=[(\mu_j-V_j)/A_j]^{3/2}$
399: with $\mu_j$ the chemical potential. They yield identical results
400: for time-independent stationary states.
401: However,
402: the
403: present time-dependent
404: model can be used in the study of
405: nonequilibrium dynamics, as in the study of soliton trains.
406:
407:
408:
409:
410: For $\nu =0$, (\ref{e})
411: can be reduced to an effective
412: one-dimensional form by considering
413: solutions of the type
414: $\psi_j({\bf r},t)=\sqrt N_j \phi_j(z,t)\psi_j^{(0)}( \rho)$
415: where
416: \begin{eqnarray}\label{wfx}
417: |\psi_j^{(0)}(\rho)|^2&\equiv&
418: {\frac{m\omega}{\pi\hbar}}\exp\left(-\frac{m
419: \omega
420: \rho^2}{\hbar}\right),
421: \end{eqnarray}
422: with
423: $m=3m_j $
424: corresponds to the
425: ground state wave function in the radial trap alone
426: in the absence of
427: nonlinear interactions. Here to have an algebraic simplification we have
428: taken the masses of two types of fermions to be equal ($m_1=m_2$).
429: These wave functions satisfy
430: \begin{eqnarray}
431: -\frac{\hbar^2}{2m}\nabla_\rho ^2\psi_j^{(0)}
432: +
433: \frac{1}{2}m\omega^2\rho^2
434: \psi_j^{(0)}&=&\hbar\omega
435: \psi_j^{(0)},
436: \end{eqnarray}
437: with normalization
438: $2\pi \int_{0}^\infty |\psi_j^{(0)}(\rho)|^2 \rho d\rho=1.$
439: Now the dynamics is carried by $ \phi_j(z,t)$ and the radial dependence is
440: frozen in the ground state $\psi_j^{(0)}(\rho)$.
441: In the quasi-one-dimensional cigar-shaped
442: geometry the
443: linear fermionic probability densities are given by
444: $|\phi_j(z,t)|^2$.
445:
446:
447:
448: Averaging over the radial mode $\psi_i^{(0)}(\rho)$,
449: i.e., multiplying
450: (\ref{e})
451: by $\psi_i^{(0)*}(\rho)$
452: and integrating over $\rho$, we obtain the following one-dimensional
453: dynamical equations \cite{mix}:
454: \begin{eqnarray}\label{i}
455: \biggr[ i \hbar\frac{\partial
456: }{\partial t}
457: +\frac{\hbar^2}{2m}\frac{\partial^2}{\partial z^2}
458: %\nonumber \\
459: -F_{jj}|
460: \phi_j|^{4/3}
461: - F_{jk}| \phi_k|^2
462: \biggr] \phi_j(z,t)=0, \nonumber \\
463: \end{eqnarray}
464: where
465: \begin{eqnarray}
466: F_{jk}=g_{12}N_k\frac{\int_0^\infty|\psi_j^{(0)}|^2|\psi_k^{(0)}|^2
467: \rho d\rho}
468: {\int_0^\infty|\psi_j^{(0)}|^2\rho d\rho}=
469: g_{12}{\frac{N_km\omega}{2\pi\hbar}}, \nonumber
470: \end{eqnarray}
471: \begin{eqnarray}
472: F_{jj}=A_jN_j^{2/3}
473: \frac{\int_0^\infty|\psi_j^{(0)}|^{2+4/3}\rho
474: d\rho}{\int_0^\infty|\psi_j^{(0)}|^2\rho
475: d\rho} =
476: {\frac{3A_j}{5}}\left[
477: \frac{N_jm\omega}{\pi \hbar} \right]^{2/3}.\nonumber
478: \end{eqnarray}
479: In (\ref{i})
480: the normalization
481: is given by $\int_{-\infty}^\infty |\phi_j(z,t)|^2
482: dz = 1$. In these equations we have set the anisotropy parameter
483: $\nu=0$
484: to remove the axial trap and thus to generate axially-free
485: quasi-one-dimensional solitons.
486:
487:
488:
489:
490: To reduce three-dimensional equations (\ref{i})
491: to a dimensionless form,
492: following \cite{fbs2},
493: we consider variables
494: $\tau=t \omega/2$, $y=z /l$, ${\varphi}_i= \sqrt l \phi_i$, with
495: $l=\sqrt{\hbar/( \omega m)}$, while (\ref{i}) becomes
496: \begin{eqnarray}\label{m} \biggr[ i\frac{\partial
497: }{\partial \tau}
498: +\frac{\partial ^2}{\partial y^2}
499: -
500: N_{jj}
501: \left|{{\varphi}_j}\right|^{4/3}
502: +N_{jk}
503: \left|{{\varphi}_k}\right|^2
504: \biggr]{\varphi}_{{j}}({y},\tau)=0,
505: \nonumber \\
506: \end{eqnarray}
507: where
508: $N_{jj}=9(6\pi N_j)^{2/3}/5, $ and
509: $N_{jk}=12 |a_{12}|N_k/l$. Here we employ
510: a negative $a_{12}$ corresponding to
511: attraction, and
512: $\int_{-\infty}^\infty |\varphi_j(y,\tau)|^2 dy =1 .$
513: In (\ref{m})
514: a sufficiently strong
515: attractive fermion-fermion coupling
516: $N_{jk}|\varphi_k|^2 (j\ne k)$ can overcome the Pauli
517: repulsion $N_{jj}|\varphi_j|^{4/3}$
518: and form bright solitons.
519:
520:
521:
522:
523:
524: For the usual Gross-Pitaevskii with the cubic nonlinearity, the
525: reduction of the full
526: three-dimensional
527: equation to its one-dimensional counterpart was performed, in different
528: forms, by
529: many
530: authors \cite{Luca}.
531: The nonlinearity with 4/3 power in (\ref{m}) was obtained starting from
532: a similar nonlinearity in a three-dimensional DFG. A strict
533: one-dimensional (two-dimensional) DFG will give rise to a quintic
534: (cubic)
535: nonlinearity. Whatever be the nonlinearity in the intraspecies fermions,
536: bright fermionic solitons will be generated, provided that the
537: interspecies attraction is strong enough to overcome the intraspecies
538: repulsion due to Pauli blocking. However, in this paper we shall
539: consider only the Pauli-blocking nonlinearity with the 4/3 power as in
540: (\ref{m}).
541:
542:
543:
544:
545:
546: The two coupled equations (\ref{m}) could be simplified for $N_1=N_2=N$,
547: while
548: $\varphi_1=\varphi_2 \equiv \varphi$, and
549: these equations reduce to the following single
550: equation
551: \begin{eqnarray}\label{o} \biggr[ i\frac{\partial
552: }{\partial \tau}
553: + \frac{\partial ^2}{\partial y^2}
554: - \beta
555: \left|{\varphi}\right|^{4/3}
556: +\gamma
557: \left|{{\varphi}}\right|^2
558: \biggr]{\varphi}({y},\tau)=0,
559: \end{eqnarray}
560: where $\beta=N_{11}=N_{22}$ and $\gamma=N_{12}=N_{21}$. Equation
561: (\ref{o}) maintains the essential features of (\ref{m}), e.g., a
562: quadratic nonlinear attraction and a Pauli blocking repulsion.
563: The one-dimensional
564: equation (\ref{o}) is quite similar in structure to the NLS equation
565: (\ref{nls}) apart for the Pauli-blocking repulsive term $\beta$.
566: For a small $\beta$ and large
567: $\gamma$
568: the solitons of
569: (\ref{nls}) survive in (\ref{o}). However, they disappear in the
570: opposite limit of large $\beta$ and small $\gamma$.
571:
572:
573:
574: \section{Modulational Instability}
575:
576: \subsection{Symmetric Case ($N_1=N_2$)}
577:
578: We find that
579: (\ref{o}) allows a constant-amplitude
580: solution which
581: exhibits modulational instability leading
582: to a modulation of
583: the solution.
584: We perform a
585: stability analysis of this
586: solution
587: and study the
588: possibility of generation
589: of solitons by modulational instability
590: in the symmetric case.
591: We consider the
592: constant-amplitude solution \cite{1}
593: \begin{equation}\label{so}
594: \varphi_0=A_0
595: \exp(i\delta) \equiv A_0 \exp [i(\gamma A_0^2 \tau -\beta A_0^{4/3}
596: \tau)]
597: \end{equation}
598: of (\ref{o}),
599: where $A_0$ is the constant amplitude and
600: $\delta$ a phase.
601: The time evolution of solution (\ref{so})
602: maintains the constant amplitude $A_0$ but acquires an amplitude-dependent
603: phase. Now we study if this
604: solution is stable
605: against small perturbations by performing a linear stability analysis.
606:
607: We consider a small perturbation of the constant-amplitude
608: solution
609: (\ref{so}) given by:
610: \begin{equation}\label{per}
611: \varphi=(A_0+ A)\exp(i \delta),
612: \end{equation}
613: where $A=A(y, \tau)$ is the
614: small perturbation.
615: Substituting the
616: perturbed solution (\ref{per})
617: in (\ref{o}), and for small perturbations retaining
618: only the linear terms in $A$ we get
619: \begin{eqnarray}\label{p} i\frac{\partial A
620: }{\partial \tau}
621: + \frac{\partial ^2 A}{\partial y^2}
622: - \frac{2}{3} \beta A_0^{4/3}(A+A^*)
623: +\gamma A_0^2 (A+A^*)=0.\nonumber \\
624: \end{eqnarray}
625:
626: We consider the complex plane-wave
627: perturbation
628: \begin{equation}
629: A(y,\tau)=
630: {\cal A}_1 \cos (K\tau -\Omega y)+i {\cal A}_2 \sin (K\tau -\Omega y)
631: \end{equation}
632: in
633: (\ref{p}), where ${\cal A}_1$ and ${\cal A}_2$ are the amplitudes of the
634: real and imaginary parts, respectively, and
635: $K$ is a frequency parameter and
636: $\Omega$ a wave number. Then separating the
637: real and imaginary parts we get
638: \begin{eqnarray}
639: -{\cal A}_1 K & =& {\cal A}_2 \Omega^2,\\
640: -{\cal A}_2K &=& {\cal A}_1 \Omega^2 -2\gamma A_0^2
641: {\cal A}_1+\frac{4}{3}\beta A_0^{4/3}{\cal A}_1,
642: \end{eqnarray}
643: and
644: eliminating ${\cal A}_1$ and ${\cal A}_2$ we obtain the dispersion
645: relation
646: \begin{equation}\label{di}
647: K=\pm \Omega \left[\Omega^2-({2\gamma A_0^2-\frac{4}{3}\beta
648: A_0^{4/3}})\right]^{1/2}.
649: \end{equation}
650: The constant-amplitude solution (\ref{so}) is stable if perturbations
651: at any wave number $\Omega$ do not grow with time. This is true as long
652: as frequency $K$ is real. From (\ref{di}) we find that
653: $K$ remains real for any $\Omega$ provided that
654: $2\gamma A_0^2 < 4\beta A_0^{4/3}/3$ or $\gamma A_0^{2/3}
655: < 2\beta /3$. However, $K$ can become imaginary for
656: $\gamma A_0^{2/3}
657: > 2\beta /3$ and the plane-wave perturbations can grow exponentially
658: with time $\tau$. This is the domain of modulational instability
659: of a constant-intensity solution \cite{1}. The perturbation then grows
660: exponentially with the intensity given by the growth rate or the
661: modulational instability gain $g(\Omega)$
662: defined by
663: \begin{equation}
664: g(\Omega)\equiv 2 \Im (K)
665: = 2|\Omega|\biggr[ 2\gamma A_0^2-\frac{4}{3}\beta
666: A_0^{4/3}-\Omega^2 \biggr]^{1/2},
667: \end{equation}
668: where $\Im$ denotes imaginary part.
669: The presence of modulational instability is closely
670: connected
671: with the appearance of a bright soliton \cite{1}.
672: Localized bright
673: solitons are possible only when the constant-amplitude solution is
674: unstable.
675:
676:
677: \subsection{Asymmetric Case ($N_1\ne N_2$)}
678:
679: Now we consider the possibility of modulational instability of a similar
680: constant-intensity solution in coupled
681: equations (\ref{m}). We consider the
682: constant-amplitude solutions
683: \begin{eqnarray}
684: \varphi_{j0}=A_{j0}\exp(i\delta_j)\equiv A_{j0}\exp[i\tau(N_{jk}
685: A_{j0}^2
686: -N_{jj} A_{j0}^{4/3} )], \nonumber
687: \end{eqnarray}
688: of (\ref{m}), where $A_{j0}$ is the amplitude and $\delta_j$ a phase
689: for component $j$. The constant-amplitude solution develops an
690: amplitude dependent phase on time evolution.
691: We consider a small perturbation $A_j\exp(i\delta_j)$ to
692: these solutions via
693: \begin{eqnarray}
694: \varphi_j=(A_{j0}+A_j)\exp(i\delta_j),
695: \end{eqnarray}
696: where $A_j=A_j(y,\tau)$.
697: Substituting this perturbed solution in (\ref{m}),
698: and for small perturbations retaining
699: only the linear terms in $A_j$ we get
700: \begin{eqnarray}\label{q}
701: i\frac{\partial A_j
702: }{\partial \tau}
703: + \frac{\partial ^2 A_j}{\partial y^2}
704: - \frac{2}{3} N_{jj} A_{j0}^{4/3}(A_j+A_j^*)+
705: \end{eqnarray}\begin{eqnarray}
706: +N_{jk}
707: A_{k0}A_{j0} (A_k+A_k^*)=0, \quad j\ne k.\nonumber
708: \end{eqnarray}
709: We consider the complex plane-wave
710: perturbation
711: \begin{equation}
712: A_j(y,\tau)=
713: {\cal A}_{j1} \cos (K\tau -\Omega y)+i {\cal A}_{j2} \sin (K\tau -\Omega
714: y)\nonumber
715: \end{equation}
716: in
717: (\ref{q}) for $j=1,2$, where ${\cal A}_{j1}$ and ${\cal A}_{j2}$
718: are the amplitudes for the real and imaginary parts, respectively, and
719: $K$ and $\Omega$ are frequency and wave numbers.
720: Then separating the real and imaginary parts we
721: get
722: \begin{eqnarray}\label{p1}
723: -{\cal A}_{11}K&=&{\cal A}_{12}\Omega^2 \\
724: -{\cal
725: A}_{12}K&=&{\cal
726: A}_{11}\Omega^2-2N_{12}A_{10}A_{20}{\cal A}_{21}+\frac{4}{3}N_{11}
727: A_{10}^{4/3}{\cal A}_{11}, \nonumber \\ \label{p2}
728: \end{eqnarray}
729: for $j=1$, and
730: \begin{eqnarray}\label{p3}
731: -{\cal A}_{21}K&=&{\cal A}_{22}\Omega^2 \\
732: -{\cal
733: A}_{22}K&=&{\cal
734: A}_{21}\Omega^2-2N_{21}A_{10}A_{20}{\cal A}_{11}+\frac{4}{3}N_{22}
735: A_{20}^{4/3}{\cal A}_{21}, \nonumber \\ \label{p4}
736: \end{eqnarray}
737: for $j=2$.
738: Eliminating ${\cal A}_{12}$ between (\ref{p1}) and
739: (\ref{p2}) we
740: obtain
741: \begin{eqnarray}\label{p5}
742: {\cal A}_{11}[K^2-\Omega^2( \Omega^2+4N_{11} A_{10}^{4/3}]=
743: 2{\cal A}_{21}N_{12}A_{10}A_{20} \Omega^2,
744: \nonumber \\ \end{eqnarray}
745: and eliminating ${\cal A}_{22}$ between (\ref{p3}) and
746: (\ref{p4}) we
747: obtain
748: \begin{eqnarray}\label{p6}
749: {\cal A}_{21}[K^2-\Omega^2( \Omega^2+4N_{22}A_{20}^{4/3}]=
750: 2{\cal A}_{11}N_{21}A_{10}A_{20} \Omega^2.
751: \nonumber \\ \end{eqnarray}
752: Eliminating ${\cal A}_{11}$ and ${\cal A}_{21}$
753: from (\ref{p5}) and (\ref{p6}),
754: finally, we obtain the following
755: dispersion relation
756: \begin{eqnarray}\label{p7}
757: K^2= \pm \Omega\biggr[ \biggr(\Omega^2+
758: \frac{2}{3}N_{11}A_{10}^{4/3}
759: +
760: \frac{2}{3}N_{22}A_{20}^{4/3}\biggr) \pm
761: \end{eqnarray}\begin{eqnarray}
762: \pm \biggr\{\frac{4}{9} \left(
763: N_{11}A_{10}^{4/3} -
764: N_{22}A_{20}^{4/3} \right)^2
765: +
766: 4N_{12}N_{21}
767: A_{10}^2A_{20}^2 \biggr\}^{1/2} \biggr]^{1/2}.\nonumber
768: \end{eqnarray}
769:
770:
771: For stability of the plane-wave perturbation, $K$ has to be
772: real. For any
773: $\Omega$ this happens for
774: \begin{eqnarray}
775: \biggr(N_{11}A_{10}^{4/3}
776: +N_{22}A_{20}^{4/3}\biggr)^2 \nonumber
777: \end{eqnarray} \begin{eqnarray}
778: > \left(
779: N_{11}A_{10}^{4/3} -
780: N_{22}A_{20}^{4/3} \right)^2+
781: 9N_{12}N_{21}
782: A_{10}^2A_{20}^2, \nonumber
783: \end{eqnarray}
784: or for $N_{12}N_{21}A_{10}^{2/3}A_{20}
785: ^{2/3}< 4 N_{11}N_{22}/9.$
786: However, for
787: $\-N_{12}N_{21}A_{10}^{2/3}A_{20}^{2/3}>
788: 4 N_{11}N_{22}/9$
789: \cite{shuk}, $K$ can become imaginary and the plane-wave perturbation can
790: grow exponentially with time. This is the domain of modulational
791: instability of a constant-intensity solution signalling the possibility
792: of coupled fermionic bright soliton to appear. In the symmetric case
793: $N_1=N_2$ and $A_{10}=A_{20}=A_0$, consequently, $N_{11}=N_{22}=\beta$
794: and $N_{12}=N_{21}=\gamma$
795: and we recover the condition of modulational instability
796: $\gamma A_0^{2/3}> 2 \beta/3$ derived in section 3.1.
797:
798:
799:
800: \section{Variational Analysis}
801:
802:
803: \subsection{Symmetric Case ($N_1=N_2$)}
804:
805: Next we present a
806: variational analysis of (\ref{o}) based on the
807: Gaussian trial wave function \cite{and}
808: \begin{equation}\label{v}
809: \varphi_v(y,\tau)=A\exp\biggr[ -\frac{y^2}{2R^2(\tau)}+\frac{i}{2}
810: b(\tau)y^2+ic(\tau) \biggr],
811: \end{equation}
812: where $A$ is the amplitude, $R$ is the width, $b$ the chirp, and $c$
813: the phase.
814: The Lagrangian density for (\ref{o})
815: is the one-term version of
816: (\ref{yy}), e.g.,
817: \begin{eqnarray}
818: {\cal L}=\frac{i}{2}\hbar \biggr[\varphi
819: \frac{\partial \varphi^*
820: }{\partial t } - \varphi ^* \frac{\partial \varphi
821: }{\partial t }
822: \biggr]+ \biggr|\frac{\partial \varphi}{\partial y} \biggr|^2
823: -\frac{1}{2}
824: \gamma n ^2
825: + \frac{3}{5}\beta n^{
826: 5/3},\nonumber
827: \end{eqnarray}
828: which is evaluated with this
829: trial function and the effective
830: Lagrangian $L
831: =\int_{-\infty}^\infty {\cal L}(\varphi_v)dy$ becomes
832: \begin{eqnarray}
833: L =\frac{A^2R\sqrt \pi}{2}\biggr(
834: \frac{6\sqrt 3}{5\sqrt 5}
835: \beta A ^{4/3}+2\dot c -
836: \frac{\gamma }{\sqrt 2} A ^2
837: +
838: \end{eqnarray}
839: \begin{eqnarray}
840: +\frac{R^2\dot b}{2} +
841: \frac{1}{R^2}
842: +b^2R^2
843: \biggr),\nonumber
844: \end{eqnarray}
845: where the overhead dot denotes time derivative.
846: The variational Lagrange
847: equations %\cite{gold}
848: \begin{equation}\label{la}
849: \frac{d}{dt}\frac{\partial L}{\partial \dot q}= \frac{\partial
850: L}{\partial q},
851: \end{equation}
852: where $q$ stands
853: for $c$, $A$, $R$ and $b$
854: can then be written as
855: \begin{eqnarray}\label{con}
856: A^2 R = \frac{1}{\sqrt \pi}=\mbox{constant}.
857: \end{eqnarray}
858: \begin{eqnarray}
859: \dot b&=& -\frac{2}{R^4}-2b^2+2\sqrt 2\gamma \frac{A^2}{R^2}
860: -4\frac{\sqrt 3}{\sqrt 5}\beta \frac{A^{4/3}}{R^2}-\frac{4\dot
861: c}{R^2},\label{x}\\
862: 3\dot b&=& \frac{2}{R^4}-6b^2+\sqrt 2 \gamma \frac{A^2}{R^2}
863: -\frac{12}{5}\frac{\sqrt 3}{\sqrt 5}\beta \frac{A
864: ^{4/3}}{R^2}-\frac{4\dot c}{R^2},\label{y}\\
865: \dot R &=& 2 R b.\label{z}
866: \end{eqnarray}
867: The constant in (\ref{con}) is fixed by the normalization condition.
868: Eliminating $\dot c$ from
869: (\ref{x}) and (\ref{y}) we obtain
870: \begin{eqnarray} \label{eli}
871: \dot b = \frac{2}{R^4}-2b^2-\frac{1}{\sqrt 2} \gamma\frac{A ^2}{R^2}
872: +\frac{4}{5}\frac{\sqrt 3}{\sqrt 5} \beta \frac{A^{4/3}}{R^2}.
873: \end{eqnarray} The use of
874: (\ref{con}), (\ref{z}) and (\ref{eli})
875: leads to the following
876: differential equation for the width $R$:
877: \begin{eqnarray} \label{min}
878: \frac{d^2 R}{d\tau^2}& =&
879: \biggr(
880: \frac{4}{R^3}- \frac{\gamma}{R^2} \sqrt\frac{2}{\pi}
881: +\frac{1}{R^{5/3}}\frac{8\beta \sqrt
882: 3}{5\pi^{1/3}
883: \sqrt5} \biggr)\\
884: &=&-\frac{d}{dR}\biggr[ \frac{2}{R^2}-\frac{\gamma}{R}
885: \sqrt\frac{2}{\pi} +
886: \frac{1}{R^{2/3}}\frac{12\beta \sqrt
887: 3}{5\pi^{1/3}
888: \sqrt5}
889: \biggr].
890: \end{eqnarray}
891: The quantity in the square bracket is the effective potential of the
892: equation of motion. Small oscillation around a stable configuration is
893: possible when there is a minimum in this potential.
894: The variational result for width $R$ follows by setting the
895: right hand side of
896: (\ref{min}) to zero corresponding to a minimum in
897: this effective potential, from which the variational
898: profile for the soliton can be obtained \cite{and}.
899:
900: \subsection{Asymmetric Case ($N_1\ne N_2$)}
901:
902: The above variational analysis can be extended to the asymmetric
903: case. However, the algebra becomes quite involved if we take a general
904: variational trial wave function with chirp and phase parameters. As we
905: are interested mostly in the density profiles, we consider the following
906: normalized Gaussian trial wave function for fermion component $j$ of
907: (\ref{m})
908: \begin{equation}\label{vx}
909: \varphi_{vj}= \sqrt{\frac{1}{R_j(\tau)\sqrt \pi}} \exp\biggr[
910: -\frac{y^2}{2R_j^2(\tau)}
911: \biggr], \quad j=1,2.
912: \end{equation}
913: Using essentially the Lagrangian density (\ref{yy}) in this case we
914: obtain the following effective Lagrangian
915: \begin{eqnarray}
916: L=-\frac{1}{\sqrt \pi}\frac{N_{jk}N_j}{\sqrt{R_1^2+R_2^2}}
917: +\sum_{j=1}^2\frac{N_j}{2}\biggr(\frac{6\sqrt 3}{5 \sqrt
918: 5\pi^{1/3}}\frac{N_{jj}}{R_j^{2/3}} +\frac{1}{R_j^2}
919: \biggr),\nonumber
920: \end{eqnarray}
921: with $j\ne k =1,2$.
922: The variational Lagrange equations (\ref{la}) for $R_1$ and $R_2$ now
923: become
924: \begin{eqnarray}\label{cp}
925: \frac{4}{R_j^3}-\frac{4N_{jk}}{\sqrt \pi}\frac{R_j}{(R_1^2+R_2^ 2)^{3/2}}
926: +\frac{8\sqrt 3}{5 \sqrt 5\pi ^{1/3}}
927: \frac{N_{jj}}{R_j^{5/3}}=0.
928: \end{eqnarray}
929: Equations (\ref{cp}) can be solved for variational widths $R_j$ and
930: consequently the variational profile of the wave functions obtained
931: from (\ref{vx}). When $N_1=N_2$, in (\ref{cp}) $N_{jj}= \beta$,
932: $N_{jk}=N_{kj}=\gamma$ and $R_1=R_2=R$; and in this case
933: (\ref{cp}) yields the same
934: variational widths as from the result obtained in the
935: symmetric case in
936: section 4.1 given by (\ref{min}); e. g.,
937: \begin{equation}\label{sim}
938: \frac{4}{R^3}-\frac{\gamma}{R^2}\sqrt
939: {\frac{2}{\pi}}+\frac{1}{R^{5/3}}\frac{8 \beta\sqrt 3}{5
940: \pi^{1/3}\sqrt 5}=0.
941: \end{equation}
942:
943: \section{Numerical Results}
944:
945:
946:
947:
948: We solve
949: (\ref{m}) for bright solitons
950: numerically using a time-iteration
951: method based on the Crank-Nicholson discretization scheme
952: \cite{sk1}. We discretize the coupled partial
953: differential equations (\ref{m})
954: using time step $0.0002$ and space step $0.015$ in the domain
955: $-8<y<8$. The second derivative in $y$ is discretized by a three-point
956: finite-difference rule and the first derivative in $\tau$ by a two-point
957: finite-difference rule. We perform a time evolution of (\ref{m})
958: introducing an harmonic oscillator potential $y^2$ in it
959: and setting the nonlinear terms to
960: zero, and
961: starting with the eigenfunction of the harmonic
962: oscillator problem: $\varphi_i(y,\tau)
963: =
964: \pi^{-1/4}\exp(-y^2/2)\exp(-i\tau).$
965: The extra
966: harmonic oscillator potential, set equal to zero
967: in the end,
968: only aids in starting the time evolution
969: with an exact analytic form. With this initial solution we perform time
970: evolution of (\ref{m}).
971: During the time evolution the nonlinear
972: terms are switched on slowly and the harmonic oscillator potential is
973: switched off slowly and
974: the time evolution continued to obtain the final converged
975: solutions.
976: In addition to solving the coupled equations
977: (\ref{m}), we also solved the single equation (\ref{o}) in the symmetric
978: case: $N_1=N_2$.
979: For given values of $N_1$ and $N_2$, solitons can be obtained for
980: $|a_{12}|$ above a certain value.
981: In the coupled-channel case solitons are easily obtained for a
982: smaller $|a_{12}|$ when $N_1$ and $N_2$ are not very different from each
983: other.
984:
985:
986: \begin{figure}%[!ht]
987:
988: \begin{center}
989: \includegraphics[width=.78\linewidth]{fig1.ps}
990: \end{center}
991:
992: \caption{The solitons $|\phi_j(z)|$ of (\ref{m})
993: vs. $z$ for
994: $N_1=40$, $N_2=60$, $a_{12}=-300 $ nm, while
995: $N_{11} \approx 149 $, $N_{12}= 216$, $N_{21}= 144$, and
996: $N_{22}\approx 195$.
997: The corresponding variational solutions $\phi_{jv}(z)$ calculated from
998: the coupled equations (\ref{cp})
999: are also shown.
1000: } \end{figure}
1001:
1002:
1003: In our numerical study we take $l=1$ $\mu$m and
1004: consider a DFFM consisting of two electronic states of
1005: $^{40}$K atoms. This corresponds to a radial
1006: trap of frequency $\omega= \hbar/(l^2m) \approx 2\pi \times 83$ Hz.
1007: Consequently, the unit of
1008: time is $2/\omega \approx 4$ ms.
1009: For another fermionic atom the $\omega$ value
1010: gets changed accordingly for $l=1$ $\mu$m.
1011:
1012:
1013: \subsection{Single Soliton}
1014:
1015:
1016:
1017:
1018:
1019:
1020: From a solution of coupled equations (\ref{m}) we find that these
1021: equations
1022: permit solitonic solutions provided that $|a_{12}|$ is larger than a
1023: certain threshold value consistent with the analysis of section 3.
1024: First we solve (\ref{m}) for
1025: $N_1=40 , N_2= 60$,
1026: and $a_{12}=-300$ nm. In this case no solitons are allowed for
1027: $a_{12}=-290$ nm. The solitonic solution suddenly appears as $|a_{12}|$
1028: increases past $290$ nm.
1029: The solitons in this case are shown in figure 1, where we also
1030: plot the variational solutions of coupled equations (\ref{cp}).
1031: In this case the nonlinearity parameters are
1032: $N_{11} \approx 149 $, $N_{12}= 216$,
1033: $N_{21}= 144$, and
1034: $N_{22}\approx 195$.
1035: Substituting these
1036: values of the nonlinearities in (\ref{cp}) and solving we obtain
1037: the variational
1038: widths $R_1 \approx 0.043$ and $R_2 \approx 0.050$.
1039: From (\ref{v}) and (\ref{con}) we then obtain the
1040: variational
1041: soliton profiles
1042: $|\phi_{v1}(z)| \approx 3.62\exp(-270z^2)$ and
1043: $|\phi_{v2}(z)| \approx 3.36\exp(-200z^2)$.
1044: From figure 1 we find that the
1045: variational results agree
1046: well with the numerical solutions of the coupled equations.
1047: In this case
1048: we also found the
1049: variational result in the symmetric case ($N_1= N_2= 50$)
1050: using (\ref{sim}),
1051: which lies close to the above two variational solutions.
1052:
1053:
1054:
1055:
1056: \begin{figure}%[!ht]
1057:
1058: \begin{center}
1059: \includegraphics[width=.58\linewidth]{fig2a.ps}
1060: \includegraphics[width=.58\linewidth]{fig2b.ps}
1061: \end{center}
1062:
1063: \caption{ The propagation of fermionic
1064: solitons (a) $|\phi_1(z,t)|$ and (b) $|\phi_2(z,t)|$ of figure 1
1065: vs. $z$ and $t$ for $N_1=40$ and $N_2=60$. At
1066: $t=100$ ms (marked by arrows) the bright solitons are set into small
1067: breathing oscillation by suddenly changing the fermion numbers to
1068: $N_1=60$ and $N_2=40$.}
1069: \end{figure}
1070:
1071:
1072:
1073:
1074: To test the robustness of these solitons we
1075: inflicted different perturbations on them and studied the resultant
1076: dynamics numerically.
1077: First, after the formation of the solitons of figure 1
1078: with $N_1=40$ and $N_2=60$ we
1079: suddenly changed the particle numbers to $N_1=60$ and $N_2=40$
1080: at time $t= 100$ ms.
1081: This corresponds to a sudden
1082: change
1083: of nonlinearities from $N_{11}\approx 149 $, $N_{12}=
1084: 216$, $N_{21}=
1085: 144$, and $N_{22} \approx 195$ to
1086: $N_{11}\approx 195 $, $N_{12}= 144$, $N_{21}= 216$, and
1087: $N_{22} \approx 149$.
1088: The resultant dynamics is
1089: shown in figures 2a and 2b.
1090: Due to
1091: the sudden change in nonlinearities, the fermionic bright
1092: solitons are set into stable non-periodic small-amplitude
1093: breathing oscillation. This demonstrates the robustness of the solitons.
1094:
1095:
1096:
1097:
1098:
1099:
1100: After the formation of the solitons of figure 1 with $N_1=40$ and
1101: $N_2=60$
1102: we
1103: suddenly changed the interspecies scattering length $a_{12}$ from
1104: $-300$
1105: nm
1106: to $-330$ nm
1107: at time $t= 100$ ms.
1108: This can be realized by manipulating a background
1109: magnetic field near a fermion-fermion Feshbach resonance \cite{fesh}
1110: This corresponds to a sudden
1111: change
1112: of nonlinearities from $N_{11}\approx 149 $, $N_{12}=
1113: 216$, $N_{21}=
1114: 144$, and $N_{22} \approx 195$ to
1115: $N_{11}\approx 149 $, $N_{12}\approx 238$, $N_{21}\approx 158$, and
1116: $N_{22} \approx 195$.
1117: Due to
1118: the sudden change in nonlinearities, the fermionic bright
1119: solitons are set into stable non-periodic small-amplitude
1120: breathing oscillation. Instead of plotting the soliton profile in this
1121: case we plot the root mean square (rms) size of the solitons
1122: in figure 3 which demonstrates the stable nonperiodic breathing
1123: oscillation.
1124: We also (i) gave a small displacement between the
1125: centers of these solitons and (ii) suddenly changed
1126: $\phi_1 \to
1127: 1.1 \times \phi_1$ and $\phi_2 \to
1128: 1.1 \times \phi_2$. In both cases
1129: after oscillation and dissipation the solitons
1130: continue stable propagation
1131: which shows their robust nature.
1132:
1133:
1134:
1135: \begin{figure}%[!ht]
1136:
1137: \begin{center}
1138: \includegraphics[width=.77\linewidth]{fig3.ps}
1139: \end{center}
1140:
1141: \caption{ The rms sizes
1142: of the solitons $|\phi_1(z,t)|$
1143: and $|\phi_2(z,t)|$
1144: of figure 1
1145: vs. $t$.
1146: At $t=100$ ms the
1147: solitons of figure 1
1148: are set into small
1149: breathing oscillation by suddenly changing
1150: $a_{12}$ from $-300$ nm to $-330$ nm. } \end{figure}
1151:
1152:
1153:
1154:
1155:
1156:
1157: \subsection{Soliton Train}
1158:
1159:
1160:
1161:
1162:
1163:
1164: \begin{figure}%[!ht]
1165:
1166: \begin{center}
1167: \includegraphics[width=.58\linewidth]{fig4a.ps}
1168: \includegraphics[width=.58\linewidth]{fig4b.ps}
1169: \includegraphics[width=.58\linewidth]{fig4c.ps}
1170: \end{center}
1171:
1172: \caption{Soliton trains of one, three and five solitons
1173: formed for $N_1=N_2=40$
1174: upon removing the harmonic trap $y^2$ and
1175: jumping the nonlinearities at
1176: $t=0$ from $N_{jj}\approx 149, N_{jk}\approx 144,
1177: k\ne j=1,2$ to
1178: (a) $N_{jj}\approx 149, N_{jk}\approx 192,$
1179: (b) $N_{jj}\approx 149, N_{jk}\approx 274,
1180: $ and to (c) $N_{jj}\approx 149, N_{jk} \approx 360,
1181: $ respectively, corresponding to a jump in scattering length
1182: $a_{12}$ from $-300$ nm to $-400$ nm, $-570$ nm, and $-750$ nm.
1183: For this purpose we solved the coupled equations (\ref{m}).}
1184: \end{figure}
1185:
1186:
1187:
1188: During the time evolution of (\ref{m})
1189: if the
1190: nonlinearities are changed
1191: by a small amount or changed slowly, usually one gets a single stable
1192: soliton when the final nonlinearities are appropriate.
1193: However,
1194: if the interspecies attraction is increased suddenly by a large
1195: amount by
1196: jumping $a_{12}$ from a positive (repulsive) value to a large negative
1197: (attractive) value, a
1198: soliton train is formed as in the experiment with BEC \cite{exdks1}
1199: because of
1200: modulational instability \cite{tai}.
1201:
1202: To illustrate the formation of soliton train
1203: in a fermion-fermion mixture in numerical
1204: simulation,
1205: we consider the solution of
1206: (\ref{m}) for $N_1=N_2=40$ and $a_{12}=-300$ nm
1207: with an added axial harmonic trap $y^2$
1208: corresponding to nonlinearities
1209: $N_{jj} \approx 149$ and $N_{jk}\approx 144,
1210: j\ne k=1,2$.
1211: After the formation of the solitons in the axial trap
1212: we
1213: suddenly jump the scattering length to $-400$ nm,
1214: corresponding
1215: to
1216: off-diagonal nonlinearities $N_{jk}= 192$,
1217: and also switch off the harmonic trap at time $t=0$.
1218: Although the initial value of scattering length ($a_{12}=-300$ nm)
1219: in this case corresponds to an interatomic attraction, because of strong
1220: Pauli-blocking repulsion the
1221: effective fermion-fermion interaction is repulsive in this case. However,
1222: the final
1223: value of
1224: scattering length ($a_{12}=-400$ nm) corresponds to a stronger
1225: interatomic
1226: attraction which can overcome the Pauli
1227: repulsion
1228: so that
1229: an effective fermion-fermion attraction emerges in this case
1230: which might allow the formation of
1231: soliton(s). In our numerical simulation we find that this is indeed
1232: the case. Upon a jump of the scattering length to $a_{12}=-400$ nm,
1233: we find that after
1234: some initial noise and dissipation
1235: the
1236: time evolution of (\ref{m}) generates a single stable
1237: bright
1238: soliton
1239: as shown in figure 4a.
1240:
1241:
1242:
1243: However, more solitons in the form of a soliton train can be
1244: formed for a larger jump in the scattering length.
1245: In figure 4b, from the same initial state in figure 4a we consider
1246: a
1247: larger jump of
1248: $a_{12}$ to $-570 $ nm corresponding
1249: to
1250: off-diagonal nonlinearities
1251: $N_{jk} \approx 274$.
1252: The final
1253: value of
1254: scattering length ($a_{12}=-570$ nm) in this case
1255: corresponds to a
1256: stronger
1257: interatomic
1258: attraction than considered in figure 4a
1259: which might allow the formation of
1260: soliton trains. This is verified in numerical simulation.
1261: Upon a jump of the scattering length to $a_{12}=-570$ nm,
1262: from $a_{12}=-300$ nm,
1263: we find that after
1264: some initial noise and dissipation
1265: the
1266: time evolution of (\ref{m}) generates three slowly receding
1267: bright
1268: solitons
1269: as shown in figure 4b.
1270: More solitons can be generated when the jump in the scattering length
1271: $a_{12}$ or off-diagonal
1272: nonlinearities is
1273: larger.
1274:
1275:
1276: In figure 4c we show the generation of five receding
1277: solitons of each component
1278: upon a
1279: sudden jump of the scattering length
1280: $a_{12}$ to $-750$ nm from the
1281: initial state of
1282: figure 4a. This
1283: corresponds to a jump of
1284: the off-diagonal nonlinearities to
1285: $N_{jk} \approx 360$.
1286: The formation of soliton trains from a stable initial
1287: state is due to
1288: modulational instability \cite{1}.
1289: The sudden jump in the off-diagonal nonlinearities could be
1290: effected by a jump in the interspecies scattering length
1291: $a_{12}$ obtained
1292: by
1293: manipulating a background magnetic field near a
1294: Feshbach resonance \cite{fesh}.
1295:
1296:
1297:
1298:
1299:
1300: \section{Summary}
1301:
1302:
1303: We used a coupled mean-field-hydrodynamic model for a DFFM
1304: to study the formation of bright solitons and soliton trains in a
1305: quasi-one-dimensional geometry by numerical and variational methods. We
1306: find that an attractive interspecies interaction can overcome the Pauli
1307: repulsion and form fermionic bright solitons in a DFFM. This is
1308: similar to the formation of bright solitons in a coupled boson-boson
1309: \cite{perez}
1310: and
1311: boson-fermion \cite{fbs1,fbs2} mixtures supported by interspecies
1312: interaction.
1313: We show by a linear stability analysis that when the interspecies
1314: attraction is larger than a threshold
1315: value, bright solitons can be formed due to modulational instability of a
1316: constant-amplitude solution of the nonlinear equations describing the
1317: DFFM.
1318:
1319:
1320: The
1321: stability of these solitons is demonstrated numerically through
1322: their sustained breathing oscillation initiated by a sudden small
1323: perturbation. We also illustrate the creation of soliton trains upon a
1324: sudden large jump in interspecies attraction by manipulating a background
1325: magnetic field near a Feshbach resonance \cite{fesh}
1326: resulting in a sudden jump in
1327: the
1328: off-diagonal nonlinearities.
1329: This jump transforms an effectively repulsive DFFM
1330: into an effectively attractive one, responsible for the formation of a
1331: soliton train due to modulational instability.
1332: Bright solitons and
1333: soliton trains have been created experimentally in attractive BECs in the
1334: presence of a radial trap only without any axial trap \cite{exdks1}
1335: in a similar fashion by transforming a repulsive BEC into an
1336: attractive one. In
1337: view of this, fermionic bright solitons and trains could be created in
1338: laboratory in a DFFM in a quasi-one-dimensional configuration.
1339:
1340: Here we
1341: used a set of mean-field equations for the DFFM. A proper treatment of a
1342: DFG or DFFM should be done using a fully antisymmetrized many-body Slater
1343: determinant wave function \cite{yyy1,fbs1,Mario} as in the case of
1344: scattering
1345: involving many electrons \cite{ps}. However, in view of the success of a
1346: fermionic mean-field model
1347: in studies of collapse \cite{ska}, bright \cite{fbs2}
1348: and dark \cite{fds} soliton
1349: in a DBFM and of mixing and demixing in a DFFM \cite{mix},
1350: we
1351: do
1352: not believe that the present study on bright solitons in a DFFM to be so
1353: peculiar as to have no general validity.
1354:
1355:
1356:
1357:
1358: %\ack
1359:
1360: \ack{
1361: The work is
1362: supported in part by the CNPq and FAPESP
1363: of Brazil.}
1364:
1365:
1366: \section*{References}
1367:
1368: \begin{thebibliography}{99}
1369:
1370:
1371:
1372: \bibitem{exdks1} Strecker K E {\it et al.} 2002 Nature {\bf 417}
1373: 150
1374:
1375: \bibitem{exdks2}
1376: Khaykovich L {\it et al.} 2002 {\it
1377: Science} {\bf 296} 1290
1378:
1379:
1380:
1381:
1382: %\bibitem{exdks1}K. E. Strecker, G. B. Partridge, A. G. Truscott, R. G.
1383: %Hulet, Nature (London) { 417} (2002) 150. \bibitem{exdks2}L.
1384: %Khaykovich,
1385: %F. Schreck, G. Ferrari, T. Bourdel, J. Cubizolles, L. D. Carr, Y.
1386: %Castin, C. Salomon, Science { 296} (2002) 1290.
1387:
1388: \bibitem{yuka} Dalfovo F, Giorgini S, Pitaevskii L P and
1389: Stringari S 1999 {\it Rev. Mod. Phys.} {\bf 71} 463
1390:
1391:
1392: Yukalov V I 2004 {\it Laser Phys. Lett.} {\bf
1393: 1} 435
1394:
1395: Yukalov V I and Yukalova E R
1396: {\it Laser Phys. Lett.} 2004 {\bf 1} 50
1397:
1398: Yukalov V I and Girardeau M D
1399: {\it Laser Phys. Lett.} 2005 {\bf 2} 375.
1400:
1401: Abdullaev F K,
1402: Gammal A, Kamchatnov A M and Tomio L 2005 {\it Int. J. Mod. Phys. B}
1403: \textbf{19} 3415
1404:
1405:
1406: \bibitem{exp1}DeMarco B and Jin D S 1999 {\it Science} {\bf 285} 1703
1407:
1408:
1409:
1410: \bibitem{exp2} O'Hara K M, Hemmer S L, Gehm M E, Granade S R and
1411: Thomas J E 2002 {\it Science} {\bf 298} 2179
1412:
1413:
1414: \bibitem{exp3}Schreck F, Khaykovich L, Corwin K L, Ferrari G,
1415: Bourdel T,
1416: Cubizolles J and Salomon C 2001 {\it
1417: Phys. Rev. Lett.} {\bf 87} 080403
1418:
1419: Truscott A G, Strecker K E, McAlexander W I,
1420: Partridge G B and Hulet R G 2001 {\it Science} {\bf 291} 2570
1421:
1422:
1423:
1424:
1425:
1426: \bibitem{exp4} Hadzibabic Z, Stan C A,
1427: Dieckmann K, Gupta S, Zwierlein M W, Gorlitz A and Ketterle W 2002
1428: {\it Phys. Rev. Lett.} {\bf 88} 160401
1429:
1430:
1431:
1432:
1433:
1434: \bibitem{exp5}Modugno G, Roati G, Riboli F, Ferlaino F,
1435: Brecha R J
1436: and Inguscio M 2002 {\it Science} {\bf 297} 2240
1437:
1438: Ospelkaus C, Ospelkaus S, Sengstock K and
1439: Bongs K 2006 \PRL {\bf 96} 020401
1440:
1441:
1442:
1443: \bibitem{exp5x} Roati G, Riboli F, Modugno G and Inguscio M 2002
1444: {\it Phys. Rev. Lett.} {\bf 89} 150403
1445:
1446:
1447:
1448:
1449:
1450:
1451:
1452: \bibitem{exp6}Strecker K E, Partridge G B and Hulet R G 2003
1453: {\it Phys. Rev. Lett.} {\bf 91} 080406
1454:
1455:
1456: Hadzibabic Z, Gupta S, Stan C A, Schunck C H, Zwierlein M W,
1457: Dieckmann K and Ketterle W 2003 {\it Phys. Rev. Lett.} {\bf 91} 160401
1458:
1459:
1460:
1461:
1462:
1463:
1464: \bibitem{yyy1} Molmer K 1998 {\it Phys. Rev. Lett.}
1465: {\bf 80} 1804
1466:
1467:
1468:
1469:
1470: \bibitem{yyy} Roth R 2002 {\it Phys. Rev. } A {\bf 66} 013614
1471:
1472: Roth R and
1473: Feldmeier H 2001 \jpb {\bf 34} 4629
1474:
1475: Roth R and
1476: Feldmeier H 2002 \PR A {\bf 65} 021603R
1477:
1478: Miyakawa T,
1479: Suzuki T and Yabu H 2001 \PR A {\bf 64} 033611
1480:
1481:
1482: Liu X-J and Hu H 2003 \PR A
1483: {\bf 67} 023613
1484:
1485: Vichi L, Amoruso M, Minguzzi A, Stringari S
1486: and Tosi M P 2000 {\it Eur. Phys. J. } D {\bf 11} 335
1487:
1488: Amoruso A, Meccoli I, Minguzzi A and Tosi M P 2000 {\it Eur.
1489: Phys. J. D} \textbf{8} 361
1490:
1491:
1492: Modugno M, Ferlaino F, Riboli F, Roati G,
1493: Modugno G
1494: and
1495: Inguscio M 2003 \PR A
1496: {\bf 68} 043626
1497:
1498:
1499:
1500: Jezek D M {\it et al.} 2004
1501: {\it Phys. Rev. A} {\bf 70} 043630
1502:
1503:
1504:
1505:
1506:
1507:
1508:
1509: \bibitem{capu}
1510: Capuzzi P, Minguzzi A and Tosi M P 2004
1511: {\it Phys. Rev. A } {\bf 69}
1512: 053615
1513:
1514: Capuzzi P, Minguzzi A and Tosi M P 2003
1515: {\it Phys. Rev. A } {\bf 67}
1516: 053605
1517:
1518:
1519:
1520: \bibitem{ska} Adhikari S K 2004 {\it Phys. Rev. A} {\bf 70} 043617
1521:
1522:
1523:
1524: \bibitem{vpg} P\'erez-Garc\'ia V M, Michinel H and Herrero H 1998
1525: \PR
1526: A
1527: {\bf 57} 3837
1528:
1529: Adhikari S K 2003 {\it New J. Phys.} {\bf 5}
1530: 137
1531:
1532: \bibitem{1} Kivshar Y S and Agrawal G P 2003 {\it Optical
1533: Solitons
1534: - From
1535: Fibers to Photonic Crystals} (San Diego, Academic)
1536:
1537:
1538:
1539: \bibitem{soldy} Abdullaev F K, Caputo J G, Kraenkel R A and Malomed B A
1540: 2003 \PR A {\bf 67} 013605
1541:
1542: Saito H and Ueda M 2003 \PRL {\bf 64} 040403
1543:
1544: Montesinos G D, P\'erez-Garc\'ia V M, Michinel H
1545: 2004 \PRL {\bf 92} 133901
1546:
1547: Adhikari S K \PR A {\bf 69} 063613
1548:
1549: Adhikari S K \PR E {\bf 71} 016611
1550:
1551:
1552: \bibitem{gs} Alfimov G L, Konotop V V and Salerno M 2002
1553: {Europhys. Lett.} \textbf{58} 7
1554:
1555: Louis P J Y, Ostrovskaya E A, Savage C M and Kivshar Y S 2003 \PR
1556: A
1557: \textbf{67} 013602
1558:
1559: Abdullaev F K and Salerno M 2005 \PR A {\bf 72} 033617
1560:
1561: Mayteevarunyoo T and Malomed B A 2006 \PR A
1562: \textbf{74} 033616
1563:
1564:
1565:
1566:
1567:
1568:
1569: \bibitem{fbs1} Karpiuk T {\it et al.} 2004 {\it Phys. Rev. Lett.} {\bf
1570: 93} 100401
1571:
1572: Karpiuk T, Brewczyk M and Rzaewski K 2006
1573: \PR A {\bf 73} 053602
1574:
1575: \bibitem{fbs2} Adhikari S K 2005 {\it Phys. Rev. A} {\bf 72} 053608
1576:
1577:
1578:
1579:
1580:
1581: Adhikari S K 2006 {\it Laser Phys. Lett. }
1582: {\bf 3} 553
1583:
1584:
1585:
1586: \bibitem{perez} P\'erez-Garc\'ia V M and Beitia J B 2005 {\it
1587: Phys. Rev.} A {\bf
1588: 72}
1589: 033620
1590:
1591: Adhikari S K 2005
1592: {\it Phys. Lett. A} {\bf 346} 179
1593:
1594:
1595:
1596:
1597: \bibitem{skal} Adhikari S K 2002 {\it Phys. Rev.} A {\bf 66} 013611
1598:
1599: Adhikari S K 2001 {\it Phys. Rev.} A
1600: {\bf 63} 043611
1601:
1602:
1603: \bibitem{fds}Adhikari S K 2005 \jpb {\bf 38} 3607
1604:
1605:
1606:
1607:
1608:
1609:
1610:
1611: \bibitem{bla}
1612: Adhikari S K 2006 {\it J. Low Temp. Phys.} {\bf 143} 267
1613:
1614: Adhikari S K 2006 {\it Laser Phys. Lett.}
1615: {\bf 3} 605
1616:
1617:
1618: \bibitem{mix} Adhikari S K 2006 \PR A {\bf 73}
1619: 043619
1620:
1621: Adhikari S K and Malomed B A 2006 \PR A \textbf{74}
1622: 053620
1623:
1624:
1625: \bibitem{fesh} O'Hara K M {\it et al.} 2002 {\it Phys. Rev. A} {\bf
1626: 66} 041401(R)
1627:
1628: Dieckmann K {\it
1629: et al.} 2002 {\it
1630: Phys. Rev. Lett.} {\bf 89} 203201
1631:
1632: Loftus T {\it et al.} 2002 {\it
1633: Phys. Rev. Lett.} {\bf 88} 173201
1634:
1635: Regal C A, Greiner M and Jin D S 2004 {\it
1636: Phys. Rev. Lett.} {\bf 92}
1637: 083201
1638:
1639:
1640: \bibitem{gehm}Gehm M E, Hemmer S L, Granade S R, O'Hara K M
1641: and
1642: Thomas J E 2003 {\it Phys. Rev.} A {\bf 68} 011401(R)
1643:
1644:
1645:
1646: \bibitem{kkk} Kokkelmans S J J M F, Milstein J N, Chiofalo M L, Walser
1647: R and
1648: Holland M J
1649: 2002
1650: \PR A {\bf 65} 053617
1651:
1652:
1653: \bibitem{hei}Heiselberg H 2001
1654: \PR A {\bf 63} 043606
1655:
1656: \bibitem{hulet1} Hulet R G 2006 private communication. I thank Dr.
1657: Hulet
1658: for a discussion, which clarified this point, suggesting the enhanced
1659: possibility of the formation of soliton train in a
1660: DFFM of different atoms.
1661:
1662:
1663: \bibitem{exp211} Chin C, Bartenstein M, Altmeyer A, Riedl S,
1664: Jochim S, Denschlag J H and Grimm R
1665: 2004 {\it Science} {\bf 305} 1128
1666:
1667: Partridge G B, Li W H, Kamar R I, Liao Y A and Hulet R G 2006
1668: {\it Science} {\bf 311} 503
1669:
1670:
1671:
1672: \bibitem{bcs} Bardeen J, Cooper L N and Schrieffer J R 1957
1673: {\it Phys. Rev.} {\bf 108}
1674: 1175
1675:
1676: \bibitem{BCSB} Eagles D M 1969 \PR {\bf 186} 456
1677:
1678: Leggett A J 1980 {\it J. Phys. (Paris) Colloq.} {\bf 41} C7-19
1679:
1680: Casas M {\it et al} 1994 \PR B {\bf 50} 15945
1681:
1682:
1683: \bibitem{hulet3}Strecker K E, Partridge G B and Hulet R G
1684: 2003 \PRL {\bf 91} 080406
1685:
1686: \bibitem{njp}Adhikari S K 2006 \NJP {\bf 8} 258
1687:
1688:
1689: \bibitem{hulet2} Al Khawaja U, Stoof H T C, Hulet R G, Strecker K E and
1690: Partridge
1691: G B
1692: 2002 \PRL {\bf 89} 200404
1693:
1694: \bibitem{sala2} Salasnich L, Parola A and Reatto L
1695: 2003 \PRL {\bf 91} 080405
1696:
1697:
1698: \bibitem{Luca}
1699: Salasnich L, Parola A and Reatto L 2002 {\it Phys. Rev.} A
1700: \textbf{65}
1701: 043614
1702:
1703: Salasnich L, Parola A and Reatto L 2002 {\it Phys. Rev.} A
1704: \textbf{66}
1705: 043603
1706:
1707:
1708: Salasnich L and Malomed B A 2006 {\it Phys. Rev.} A \textbf{74} 053610
1709:
1710: Khaykovich L
1711: and Malomed B A 2006 {\it Phys. Rev.} A \textbf{74} 023607
1712:
1713:
1714:
1715:
1716: \bibitem{shuk} Kourakis I {\it et al.} 2005
1717: {\it Eur. Phys. J.} B {\bf 46} 381
1718:
1719:
1720:
1721: \bibitem{and} Anderson D 1983 {\it Phys. Rev. A} {\bf 27} 3135
1722:
1723: P\'erez-Garc\'ia V M, Michinel H, Cirac J I, Lewenstein M and
1724: Zoller P 1997 \PR A {\bf 56} 1424
1725:
1726: Stoof H T C 1997 {\it J. Stat. Phys.} {\bf 87} 1353
1727:
1728:
1729: Malomed B A 2002 in {\it Progress in Optics, vol. 43, p. 71, ed. by
1730: Wolf E} (Amsterdam, North-Holland)
1731:
1732:
1733:
1734:
1735:
1736:
1737:
1738:
1739: \bibitem{sk1}
1740: Muruganandam P and Adhikari S K 2003 \jpb {\bf 36} 2501
1741:
1742:
1743: Adhikari S K 2000 \PR E {\bf 62} 2937
1744:
1745: Adhikari S K 2002 \PR A {\bf 66} 013611
1746:
1747:
1748:
1749:
1750: \bibitem{tai}Tai K, Hasegawa A and Tomita A 1986
1751: {\it Phys. Rev. Lett.} {\bf 56} 135
1752:
1753:
1754:
1755:
1756:
1757:
1758:
1759: \bibitem{Mario} Salerno M 2005 \PR A
1760: \textbf{72} 063602
1761:
1762:
1763:
1764:
1765:
1766: \bibitem{ps}
1767: Biswas P K and Adhikari S K 2000 \jpb {\bf 33} 1575
1768:
1769: Biswas P K and Adhikari S K 1998 \jpb {\bf 31} L315
1770:
1771: Tomio L and Adhikari S K 1980 \PR C {\bf 22}
1772: 28
1773:
1774: Adhikari S K and Ghosh A 1997 \JPA {\bf 30} 6553
1775:
1776:
1777: Adhikari S K 1979 \PR C {\bf 19} 1729
1778:
1779:
1780:
1781:
1782:
1783:
1784:
1785:
1786:
1787:
1788:
1789:
1790:
1791:
1792:
1793:
1794: \end{thebibliography}
1795:
1796: \end{document}
1797:
1798:
1799: