1: %\documentclass[aps,prl,twocolumn]{revtex4}
2: %\documentclass[preprint,pra,floatfix,showpacs]{revtex4}
3: %\documentclass[aps,prl,preprint]{revtex4}
4: %\usepackage{graphicx}
5: %\documentclass[aps,onecolumn,showpacs,superscriptaddress,preprintnumbers]{revtex4}
6:
7:
8: \documentclass[aps,prl, showpacs,singlecolumn]{revtex4}
9: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
10: \usepackage{amssymb}
11: \usepackage{amsmath}
12: \usepackage{graphicx}
13: \usepackage{float}
14: \usepackage{dcolumn}
15: \usepackage{bm}
16:
17: \setcounter{MaxMatrixCols}{10}
18: %TCIDATA{OutputFilter=Latex.dll}
19: %TCIDATA{Version=5.00.0.2552}
20: %TCIDATA{<META NAME="SaveForMode" CONTENT="1">}
21: %TCIDATA{LastRevised=Monday, November 28, 2005 08:45:37}
22: %TCIDATA{<META NAME="GraphicsSave" CONTENT="32">}
23:
24:
25:
26: \begin{document}
27:
28: \title{Controlling quantum state transfer in spin chain with the confined
29: field}
30: \author{Bing Chen$^{1}$, Z. Song$^{1,a}$ and C. P. Sun $^{1,2,a,b}$}
31: \affiliation{$^{1}$Department of Physics, Nankai University, Tianjin 300071, China}
32: \affiliation{$^{2}$ Institute of Theoretical Physics, Chinese Academy of Sciences,
33: Beijing, 100080, China}
34:
35: \begin{abstract}
36: As a demonstration of the spectrum-parity matching condition (SPMC) for
37: quantum state transfer, we investigate the propagation of single-magnon
38: state in the Heisenberg chain in the confined external tangent magnetic
39: field analytically and numerically. It shows that the initial Gaussian wave
40: packet can be retrieved at the counterpart location near-perfectly over a
41: longer distance if the dispersion relation of the system meets the SPMC
42: approximately.
43: \end{abstract}
44:
45: \pacs{05.40.-a,03.65.-w,42.25.Bs,73.23.Hk}
46: \maketitle
47:
48: \section{Introduction}
49: Transferring a quantum state or object from one place to another
50: is an important task in quantum information science. Recent works
51: find that the quantum spin chain can be used to act as
52: \textquotedblleft quantum data bus\textquotedblright\ to perfectly
53: transfer quantum states which has drawn
54: the attention of quantum communication community \cite%
55: {Bose1,Liying,Song,Christandle1,Christandle2,Christandle3,Bose2,Bose3,Shitao1,Shitao2}%
56: . A great advantage of this approach is that no dynamical controls are
57: needed after we prepared the quantum state (except for the detection of the
58: state at the other end of spin chain) that is only the sending and the
59: receiving parties apply gates to the system, but the part of the chain
60: interconnecting them cannot be controlled during the communication process.
61:
62: In the first proposal \cite{Bose1}, Bose considered a regular
63: one-dimensional spin chain with Heisenberg interactions which is able to
64: transfer quantum state over reasonable distance with the aid of distillation
65: process. This proposal is not the perfect scheme to transfer quantum state
66: due to the dispersion of the spin chain. This result in the fidelity for a
67: single-qubit state transfer becomes worse as the chain get longer.
68:
69: Since then, a number of interesting proposals are proposed for quantum
70: communication through spin system to improve the fidelity of the quantum
71: information transfer. One of them is to choose proper the modulation of the
72: coupling strengths between two nearest neighbor sites as suggested in \cite%
73: {Christandle1,Christandle2,Christandle3}. In such system, although
74: an arbitrary local quantum state will spread as the time evolves,
75: after a period of time the dispersed information will
76: \textquotedblleft refocus\textquotedblright\ at the receiving
77: location of the chain. So perfect states transfer can be realized.
78: Another approach is making use of gapped systems. The advantage of
79: these schemes is the intermediate spins are only virtually excited.
80: In this case, the two separated qubits are coupled and realized the
81: entanglement of two point \cite{Liying}. It ensures that the
82: transfer of single-qubit state has a very high fidelity.
83:
84: According to quantum mechanics, perfect state transfer can be
85: realized in many continuous and discrete systems. For simplicity, we
86: only consider a single-particle system with the usual spatial
87: refection symmetry (SRS). The SRS can be expressed by $[H,P]=0$,
88: where $P$ is the spatial refection
89: operator and $H$ is the Hamiltonian of the system. If the eigen values $%
90: \varepsilon _{n}$ match the parities $P_{n}$ in the following way
91: \begin{equation}
92: \varepsilon _{n}=N_{n}E_{0},P_{n}=\pm (-1)^{N_{n}}, \label{spmc}
93: \end{equation}%
94: any state $\varphi (\vec{r})$ can evolve into the reflected state
95: \begin{equation}
96: P\varphi (\vec{r})=\pm \varphi (-\vec{r})
97: \end{equation}
98: after time $\tau =(2n+1)\pi /E_{0}$, where $\vec{r}$ is the position of
99: sites in the lattice. Eq. (\ref{spmc}) is called the spectrum-parity
100: matching condition (SPMC), which can conduct us to find various artificial
101: solid state system transferring quantum states with high fidelity. It is
102: easy to find that the model of \cite{Christandle1} corresponds to the SPMC
103: for the simplest case $N_{n}=n$.
104:
105: As analyzed above, many different schemes can realize a quantum state
106: transfer between two parties by means of a spin chain: the state of the
107: leftmost qubit is transferred to the rightmost qubit after a given time,
108: depended on the dynamics of the chain, to achieve perfect state transfer
109: over arbitrary distances. It also indicates that the revival period is
110: determined by the size of the system, but not the distance between two
111: parties. So it is necessary for us to find a new model to control the
112: revival period. On the other hand, the model we considered must satisfy the
113: SPMC. The previous work \cite{Shitao2} has considered $(2N+1)$-site spin-$%
114: 1/2 $ ferromagnetic Heisenberg chain in the parabolic magnetic field, which
115: is the most simple model satisfying SPMC. Inspired by this paper, it is
116: natural for us to study other extension of the application of the theorem
117: with $N_{n}\neq n$. In this paper, we present a new example, based on the
118: SPMC and super-symmetry theory in quantum mechanics. The protocol we
119: considered is a ferromagnetic Heisenberg chain with uniform coupling
120: constant in an external magnetic field. It has been shown \cite{Shitao1,
121: Shitao2} that if the external field is parabolic, the lower spectrum is
122: equal-spacing, which is a good example to meet SPMC in the simplest case $%
123: N_{n}=n$. The super-symmetry theory in quantum mechanics \cite{SS} implies
124: that if the applied field is tangent-shaped, the lower spectrum should
125: satisfy the SPMC in the case $N_{n}=n^{2}$. In the following sections, we
126: will investigate the propagation of single-magnon state in the Heisenberg
127: chain in the confined external tangent magnetic field analytically and
128: numerically.
129:
130: \section{The scheme and model}
131: In general, the transfer of a qubit state from the location $A$ to
132: $B$ can be regarded as the following process. The initial qubit
133: state $\left\vert \psi _{A}\right\rangle =u\left\vert
134: 1\right\rangle _{A}+v\left\vert 0\right\rangle _{A}$ is prepared
135: at $A$, where $\left\vert 1\right\rangle _{A}$ and $\left\vert
136: 0\right\rangle _{A}$ denote the two-level basis of the qubit at
137: $A$. For a spin system, we have $\left\vert 1\right\rangle
138: _{A}=\left\vert \uparrow \right\rangle _{A}$\ and $\left\vert
139: 0\right\rangle _{A}=\left\vert \downarrow \right\rangle _{A}$.\ If
140: one can find an operation $U_{AB}$\ to realize
141: \begin{equation}
142: U_{AB}\left\vert \psi _{A}\right\rangle =u\left\vert 1\right\rangle
143: _{B}+e^{i\varphi _{AB}}v\left\vert 0\right\rangle _{B}
144: \end{equation}%
145: where $\varphi _{AB}$\ is the known phase for a given system, we say the
146: qubit state is transferred from $A$ to $B$ perfectly. Bose \cite{Bose1}
147: pointed out that the operation $U_{AB}$\ can be achieved in the qubit array
148: by the time evolution of the system. Especially, the perfect state transfer
149: can be performed if the system meets the SPMC. Unfortunately, such kind of
150: array needs a deliberated modulation \cite%
151: {Christandle1,Christandle2,Christandle3, Shitao2} for the couplings between
152: two qubits, which is difficult to be pre-engineered in experiment.
153: Nevertheless, if the transferred state is not a qubit state at a certain
154: site but the superposition of the single-qubit state localized in a small
155: range of coordinate space, e.g., a Gaussian wave packet (GWP) represented as
156: \begin{equation}
157: \left\vert \psi _{A}\right\rangle =C\sum_{i}\exp \left[ -\frac{1}{2}\alpha
158: ^{2}(i-N_{A})^{2}\right] (u\left\vert 1\right\rangle _{i}+v\left\vert
159: 0\right\rangle _{i})\prod\limits_{j\neq i}\left\vert 0\right\rangle _{j},
160: \label{gwp}
161: \end{equation}%
162: where $C$ is the normalization factor, it can be transferred to a
163: certain position $N_{B}$\ near perfectly \cite{Shitao2}. It
164: indicates that, for a
165: spin model with the saturated ferromagnetic state $\prod\nolimits_{j}\left%
166: \vert 0\right\rangle _{j}$ being the eigenstate, the transfer of a
167: single-magnon GWP is equivalent to that of a qubit.
168:
169: In this section, we propose a model whose spectrum structure obeys the SPMC
170: in low energy for the near perfect transfer of a GWP. Let us consider the
171: Hamiltonian of $(2N+1)$-site spin-$1/2$ ferromagnetic Heisenberg chain
172: \begin{eqnarray}
173: H &=&-J\sum_{i=-N}^{N-1}\overrightarrow{S}_{i}\cdot \overrightarrow{S}%
174: _{i+1}+\sum_{i=-N}^{N}B(i)S_{i}^{z} \notag \\
175: &=&-J\sum_{i=-N}^{N-1}\left[ \frac{1}{2}%
176: (S_{i}^{+}S_{i+1}^{-}+S_{i}^{-}S_{i+1}^{+})+S_{i}^{z}S_{i+1}^{z}\right]
177: +\sum_{i=-N}^{N}B(i)S_{i}^{z}, \label{H}
178: \end{eqnarray}%
179: where $J>0$ is the exchange coupling constant. The external magnetic field
180: has the form%
181: \begin{equation}
182: B(i)=B_{0}\left[ \tan ^{2}\left( \pi \eta i\right) +\frac{1}{2}\right] ,
183: \label{B(i)}
184: \end{equation}%
185: where $B_{0}=2\lambda J\pi ^{2}\eta ^{2}$, and $L_{eff}=1/\eta $ denotes the
186: effective length of the system, indicating the confined range of the single
187: magnon. Here $S_{i}^{x}$, $S_{i}^{y}$ and $S_{i}^{z}$ are Pauli matrices for
188: $i$-th site. Obviously, for the Hamiltonian Eq. (\ref{H}) the $z$-component
189: of the total spin%
190: \begin{equation}
191: S^{z}=\sum_{i=-N}^{N}S_{i}^{z}
192: \end{equation}%
193: is the conservative quantity of the system, i.e.,
194: \begin{equation}
195: \left[ H,S^{z}\right] =0.
196: \end{equation}%
197: Then in the invariant subspace with the fixed $S^{z}=N-1/2$, the Ising term
198: in the Hamiltonian $J\sum_{i=-N}^{N-1}S_{i}^{z}S_{i+1}^{z}$ just contributes
199: a constant to the matrix of Hamiltonian, which can be neglected. With the
200: aid of Jordan-Wigner transformation \cite{J-W} the single-magnon effective
201: Hamiltonian can be expressed as a spin-less fermion lattice model with a
202: tan-potential
203: \begin{equation}
204: H_{eff}=-\frac{J}{2}\sum_{i=-N}^{N-1}\left( a_{i}^{\dag }a_{i+1}+h.c.\right)
205: +B_{0}\sum_{i=-N}^{N}\tan ^{2}\left( \pi \eta i\right) \label{Heff}
206: \end{equation}%
207: where $a_{i}^{\dag }$ is the fermion operator at $i$-th site, $%
208: B_{0}=2\lambda J\pi ^{2}\eta ^{2}$ and we have neglected a constant for
209: simplicity. The spectrum of Eq. (\ref{Heff})\ can not be solved exactly
210: except the case of very small $N$.\ However, the corresponding continuous
211: model\ has exact solution and can give us some implications. Actually,
212: consider the Hamiltonian Eq. (\ref{Heff}) in $B_{0}=0$\ case, the
213: tight-binding spectrum is
214: \begin{equation}
215: \epsilon _{k}=-2J\cos k,
216: \end{equation}%
217: where $k=n\pi /\left( 2N+2\right) $, $n=1,2,...,2N+1$. Notice that in lower
218: energy region $k\sim 0$\ the spectrum is $\epsilon _{k}\approx $\ $%
219: -2J(1-k^{2})$\ $\sim 2Jk^{2}$, which is very close to that of free particle
220: in a continuous system, i.e., $E=p^{2}/2m$. It implies that for lower energy
221: scale, the spectrum can be obtained from the corresponding continuous model
222: approximately. According to quantum mechanics \cite{SS}, the energy spectrum
223: of the continuous model
224: \begin{equation}
225: H_{con}=2Jp^{2}+B_{0}\tan ^{2}\left( \pi \eta x\right)
226: \end{equation}%
227: is%
228: \begin{equation}
229: E_{n}=J\pi ^{2}\eta ^{2}\left( n^{2}+4\mu n+C\right) \label{En}
230: \end{equation}%
231: \ where
232: \begin{eqnarray}
233: \mu &=&\frac{1}{4}\left( \sqrt{8\lambda +1}-1\right) , \notag \\
234: C &=&J\pi ^{2}\eta ^{2}\left( \sqrt{8\lambda +1}-1\right) .
235: \end{eqnarray}%
236: The reason to choose the tan-potential is due to its rich spectrum structure
237: as $\lambda $\ varies. For $\lambda =1$, the corresponding energy spectrum is%
238: \begin{equation}
239: E_{n}=\frac{1}{2}n(n+2)B_{0},\text{ }n=1,2,\cdots .
240: \end{equation}%
241: \ which can be written as the quadratic term of $(n+1)$.\textrm{\ }On the
242: other hand, when $\lambda \gg 1$\ is taken into account, the linear term
243: becomes dominant, which corresponds to that of parabolic potential \cite%
244: {Shitao2}.
245:
246: In this paper, we mainly focus on the case of $\lambda =1$, since the
247: quadratic dispersion also meet the SPMC. For a given $L_{eff}=500$ as an
248: example, the magnitude of magnetic field is $B_{0}/\lambda =7.9\times
249: 10^{-5}J$. The system is illustrated in Fig. 1.
250: \begin{figure}[tbp]
251: \includegraphics[bb=80 236 535 532, width=8 cm, clip]{fig1.eps}
252: \caption{Schematic figure of the system with tan-potential
253: $B(x)=B_{0}\tan ^{2}(\protect\pi \protect\eta x)$ indicated by the
254: curve line. The initial GWP localizes at the point $A$. After
255: $t=2\protect\pi /B_{0}$, the state will be retrieved at $B$ as
256: time evolves.} \label{fig1}
257: \end{figure}
258: Obviously, such spectrum meets to the SPMC for the case $N_{n}=n(n+2)$.
259: According to SPMC, any initial state $\psi \left( x\right) $ with dominant
260: components in low energy scale can be evolved into $\pm \psi \left(
261: -x\right) $ at time $\tau =2(2n+1)\pi /B_{0}$ approximately. On the other
262: hand, since the saturated ferromagnetic state $\prod\nolimits_{j}\left\vert
263: 0\right\rangle _{j}$ is the ground state of the model, which is unchanged
264: during the time evolution. Then if a GWP of a single-magnon state can be
265: transferred near perfectly, this model can be employed to implement quantum
266: information transmission in solid-based device.
267:
268: In order to demonstrate the analysis of the spectrum of the system,
269: numerical simulation is performed for the system with $L_{eff}=500$. In Fig. %
270: \ref{fig2}, we show the numerical and analytical result of energy spacing $%
271: D\left( n\right) =E_{n+1}-E_{n}$\ (in the unit of $J$) as the function of
272: energy level $n$. It shows that there is indeed a liner region for lower
273: energy levels (about $n<80$), which satisfy the SPMC approximately. So if
274: the GWP can be expanded by the lower eigenstates dominantly, it can be
275: transferred near perfectly.
276: \begin{figure}[tbp]
277: \includegraphics[bb=19 343 525 744, width=7 cm, clip]{fig2.eps}
278: \caption{The energy spacing $D(n)=E_{n+1}-E_{n}$ (in the unit of $J$)
279: obtained by numerical simulation (black line) for $L_{eff}=500$ and $B_{0}=2J%
280: \protect\pi ^{2}\protect\eta ^{2}$. The red line is the plot of Eq. (\protect
281: \ref{En}). It shows that at low energy region the spectrum of single magnon
282: is quadratic in high accuracy. }
283: \label{fig2}
284: \end{figure}
285:
286: \section{Quantum transfer of GWP}
287: In this section, we show some numerical examples of quantum state
288: transmission in the spin chain with the uniform coupling strength
289: $-J<0$, controlled in the magnetic field in the form of Eq.
290: (\ref{B(i)}). We
291: consider a GWP with the center at $N_{A}$ as the initial state at\textbf{\ }$%
292: t=0$. The packet we choose must be approximately expanded by the energy
293: levels, which satisfy the SPMC.
294: \begin{equation}
295: \left\vert \psi (N_{A},0)\right\rangle =C\sum_{i=-N}^{N}\exp \left[ -\frac{1%
296: }{2}\alpha ^{2}(i-N_{A})^{2}\right] \left\vert i\right\rangle
297: \end{equation}%
298: where $\left\vert i\right\rangle =\left\vert 1\right\rangle
299: _{i}\prod\nolimits_{j\neq i}\left\vert 0\right\rangle _{j}$ denotes the
300: state with a single spin flip at $i$-th site. The factor $\alpha $
301: determines the width of the wave-packet, $\Delta =2\sqrt{\ln 2}/\alpha $. At
302: time $t$ the state $\left\vert \psi (N_{A},0)\right\rangle $ evolves to
303: \begin{equation}
304: \left\vert \phi (t)\right\rangle =e^{-iH_{eff}t}\left\vert \psi
305: (N_{A},0)\right\rangle .
306: \end{equation}%
307: For small $\alpha $,\ $\left\vert \psi (N_{A},0)\right\rangle $\ can be
308: expanded by the eigenstates $\left\vert \varphi _{n}\right\rangle $\ of $%
309: H_{eff}$,\ which belong to the quadratic region, i.e.,
310: \begin{equation}
311: H_{eff}\left\vert \varphi _{n}\right\rangle \simeq E_{n}\left\vert \varphi
312: _{n}\right\rangle .
313: \end{equation}%
314: Thus at $\tau =2\pi /B_{0}$ the final state is
315: \begin{eqnarray}
316: \left\vert \phi (t=\tau )\right\rangle &=&e^{-iH\tau }\left\vert \psi
317: (N_{A},0)\right\rangle \notag \\
318: &\simeq &\sum_{n}e^{-iE_{n}\tau }\left\vert \varphi _{n}\right\rangle
319: \left\langle \varphi _{n}\right\vert \left. \psi (N_{A},0)\right\rangle
320: \notag \\
321: &=&\sum_{n}e^{-i\pi n(n+2)}\left\vert \varphi _{n}\right\rangle \left\langle
322: \varphi _{n}\right\vert \left. \psi (N_{A},0)\right\rangle \notag \\
323: &=&P\left\vert \psi (N_{A},0)\right\rangle \notag \\
324: &=&\left\vert \psi (-N_{A},0)\right\rangle ,
325: \end{eqnarray}%
326: which is the reflection counterpart of the initial one.
327:
328: As for quantum information, what we concern is the fidelity of qubit-state
329: transmission. Since the saturated ferromagnetic state $\prod\nolimits_{j}%
330: \left\vert 0\right\rangle _{j}$\ is the ground state of the Hamiltonian, and
331: belongs to different subspace from the single-magnon state, the fidelity of
332: quantum information is determined by the fidelity of the GWP transmission
333: \begin{equation}
334: F(t)=\left\vert \left\langle \psi (-N_{A},0)\right\vert
335: e^{-iH_{eff}t}\left\vert \psi (N_{A},0)\right\rangle \right\vert .
336: \end{equation}
337: On the other hand, the above analytical result is obtained in an approximate
338: manner. In order to investigate the accuracy of the approximation for
339: various systems, numerical simulation is employed. For this purpose, we
340: numerically solve the Schr\"{o}dinger equation for the dynamical evolution
341: and compute the fidelity of the wave-packet transmission from rightmost $%
342: N_{A}$\ of the spin chain to the leftmost $-N_{A}$.
343:
344: \begin{figure}[tbp]
345: \includegraphics[bb=17 62 531 753, width=7 cm, clip]{fig3.eps} .
346: \caption{Fidelity of the GWPs with $\Delta =24$, $18$, $12$, and $6$
347: transferred over the distance of $L=2N_{A}$ $=400$\ in the system with $%
348: L_{eff}=500$.}
349: \label{fig3}
350: \end{figure}
351:
352: \begin{figure}[tbp]
353: \includegraphics[bb=31 302 510 750, width=7 cm, clip]{fig4.eps}
354: \caption{The plot of maximal fidelity of the GWPs with $\Delta =28$, $18$
355: transferred over the distance of $L=2L_{eff}-100$ in the system with $%
356: L_{eff}\in \lbrack 500,1000]$. It shows that, as the effective length gets
357: longer, the fidelity slightly decreases if the width of GWP is taken
358: appropriately.}
359: \label{fig4}
360: \end{figure}
361: As mentioned before, the SPMC can be satisfied in the range of
362: lower spectrum. The width of GWP $\Delta $\ determines the range
363: of the levels span its wave function, as well as the fidelity of
364: the transmission. In other word, the smaller $\Delta $\ is, the
365: lower the fidelity will be. We consider the transmissions of the
366: GWP with $\Delta =24$, $18$, $12$, and $6$ over the distance
367: $L=2N_{A}$ $=400$\ in the system with $L_{eff}=500$. The
368: fidelities as function of time $t$\ are plotted in Fig.
369: \ref{fig3}. We find that the fidelities get their maxima around
370: $(2n+1)\tau $ $(n=0,1,2,...)$, which are in agreement with the
371: analytical prediction. It shows that the maximal fidelity becomes
372: high as the width of GWP increases. Then for the scheme of quantum
373: information transfer, the higher fidelity can be achieved if the
374: GWP with $\Delta \geq 24$ is employed. Interestingly, one can see
375: that there exists a regular "noise" in the plot of the fidelity.
376: This
377: phenomenon is related to the fractional revival of the wave-packet \cite%
378: {Fractional revival} and will be discussed in the further work.
379:
380: On the other hand, we also investigate the influence of the transfer
381: distance to the fidelity numerically. Numerical simulation is
382: performed to obtain the maximal fidelity $F(t)$\ of transmission of
383: GWPs with $\Delta =$\ $18$ and $28$ over the distance of $L=2N_{A}$
384: $=L_{eff}-100$ in the systems with $L_{eff}\in \lbrack 500,1000]$.
385: In Fig. \ref{fig4}, the maximal fidelity as a function of the
386: confined length $L_{eff}$\ is plotted. It shows that, as the
387: effective length gets longer, the fidelity slightly decreases if the
388: width of GWP is taken appropriately. It indicates that this scheme
389: can be a good candidate to realize quantum information transmission
390: in solid-based device.
391:
392: In the above we focus on the case of $\lambda =1$\ due to the
393: indication from the continuous system. A natural question is asked
394: whether $\lambda =1$ is the optimal for the GWP transmission in the
395: vicinity of $1$. Numerical study has been done to calculate the
396: fidelity as a function of $\lambda $\ for the propagation of a GWP
397: with $\Delta =24$ over the distance of $2N_{A}=400$\ \ in the system
398: with $L_{eff}=500$. The result is plotted in Fig. \ref{fig5}, which
399: shows that $\lambda \lesssim 1$ is approximately optimal.
400: \begin{figure}[tbp]
401: \includegraphics[bb=34 344 536 740, width=7 cm, clip]{fig5.eps}
402: \caption{Maximum fidelity of the GWPs with $\Delta =24$ transferred
403: over the distance of $L=400$ in the system with $L_{eff}=500$, as a
404: function of $\protect\lambda $. It indicates that $\protect\lambda
405: \lesssim 1 $ is approximately optimal.} \label{fig5}
406: \end{figure}
407:
408: \section{Harmonic approximation in strong field limit}
409: All above discussions are based on the tan-potential with $\lambda
410: =1$ and numerical result has demonstrated that it is the optimal
411: potential for GWP transmission. In this section, we concentrate on
412: another special case of strong field limit. From the above
413: analysis, we know that the low spectrum for arbitrary value of
414: $\lambda $ is similar to Eq. (\ref{En}), which is the combination
415: of linear and quadratic dispersions. Obviously, it is easy to find
416: that the linear dispersion is dominant when we consider the strong
417: field limit, $\lambda \gg 1$. In other word, under the condition
418: of the strong field limit, the effective spectrum for the GWP is
419: quasi-harmonic. In this case the SPMC is also satisfied and can be
420: employed to transfer quantum information \cite{Shitao2}.
421: \begin{figure}[tbp]
422: \includegraphics[bb=54 340 525 741, width=7 cm, clip]{fig6.eps}
423: \caption{The energy spectrum (in the unit of $J$) obtained by numerical
424: simulation (black line) for $L_{eff}=500$ and strong field with $B_{0}=6.33J$%
425: . The red line is the plot of $E_{n}^{\prime }=0.032Jn$. It shows that at
426: low energy region the spectrum of single magnon is linear in high accuracy.}
427: \label{fig6}
428: \end{figure}
429: \begin{figure}[tbp]
430: \includegraphics[bb=14 47 577 810, width=7.0cm, clip]{fig7.eps}
431: \caption{Fidelity of the GWPs with (a) $\Delta =6$; (b) $\Delta =4$; (c) $%
432: \Delta =2$, transferred over the distance of $L=120$ in the system with $%
433: L_{eff}=500$ and $B_{0}=6.33J$, as a function of time $t$.}
434: \label{fig7}
435: \end{figure}
436: \begin{figure}[tbp]
437: \includegraphics[bb=16 48 575 809, width=7.0cm, clip]{fig8.eps}
438: \caption{Fidelity of the GWPs with $\Delta =6$, transferred over the
439: distance of (a) $L=120$; (b) $L=200$; (c) $L=220$ in the system with $%
440: L_{eff}=500$ and $B_{0}=6.33J$, as a function of time $t$.}
441: \label{fig8}
442: \end{figure}
443: In order to demonstrate the above analysis, we perform numerical
444: simulation
445: for the system with strong field $B_{0}=$ $6.33J$ (or $\lambda =$ $%
446: 8.02\times 10^{5}$) and $L_{eff}=500$. The energy spectrum of such system
447: with obtained by exact diagonalization and the analytical result from the
448: continuous system in large $\lambda $\ limit, i.e.
449: \begin{equation}
450: E_{n}^{\prime }=2\sqrt{2\lambda }J\pi ^{2}\eta ^{2}n=0.032Jn,
451: \end{equation}%
452: are plotted in Fig. \ref{fig6}. It shows that the lower spectrum (about $%
453: n\simeq 80$) is equal-spacing in high accuracy. In Fig.
454: \ref{fig7}, we plot the fidelity as a function of time for the
455: propagation of GWP with different half-width $\Delta =2,$ $4,$ and
456: $6$ over the distance of $L=120$. It shows that, the
457: quasi-harmonic potential is more appropriate to transfer
458: wave-packet with small width and the fidelity is higher. Meanwhile
459: the recurrent period is shorter. On the other hand, we also
460: investigate the influence of the transfer distance $L$ to the
461: fidelity. From Fig. \ref{fig8} we can see that the fidelity
462: quickly decays as the transfer distance gets longer.
463:
464: \section{Conclusion}
465: We have studied analytically and numerically, the propagation of
466: single-magnon state in the Heisenberg chain in the confined
467: external tangent magnetic field with different strength. We find
468: that it can be a good model to demonstrate the SPMC for quantum
469: state transfer in spin systems. It is shown that the initial
470: Gaussian wave packet can be retrieved at the counterpart location
471: near-perfectly over a longer distance if the dispersion relation
472: of the system meets the SPMC approximately. It also shows that the
473: qubit array with appropriate external field is a suitable scheme
474: for the task of quantum information transmission in solid-state
475: system.
476:
477: This work is supported by the CNSF (grant No. 10474104), the National
478: Fundamental Research Program of China (No. 2001CB309310).
479:
480: \begin{thebibliography}{a}
481: \bibitem[a]{email} emails: songtc@nankai.edu.cn \newline
482: and suncp@itp.ac.cn
483:
484: \bibitem[b]{www} Internet www site: http://www.itp.ac.cn/\symbol{126}suncp
485:
486: \bibitem{Bose1} S. Bose, Phys. Rev. lett. \textbf{91}, 207901 (2003)
487:
488: \bibitem{Liying} Y. Li, T. Shi, B.Chen, Z. Song, C. P. Sun, Phys. Rev. A
489: \textbf{71}, 022301 (2005).
490:
491: \bibitem{Song} Z. Song and C. P. Sun, Low Temperature Physics \textbf{31},
492: 686 (2005).
493:
494: \bibitem{Christandle1} M. Christandl, N. Datta, A. Ekert and A. J. Landahl,
495: Phys. Rev. Lett. \textbf{92}, 187902 (2004)
496:
497: \bibitem{Christandle2} M.Christandle, N. Datta T. C. Dorlas, A. Ekert, A.
498: Kay and A. J. Landahl, Phys. Rev. A \textbf{71}, 032312 (2005).
499:
500: \bibitem{Christandle3} C. Albanese, M.Christandle, N. Datta and A. Ekert,
501: Phys. Rev. Lett. \textbf{93}, 230502 (2004).
502:
503: \bibitem{Bose2} Daniel Burgarth and Sougato Bose, New J. Phys. \textbf{7}
504: 135 (2005).
505:
506: \bibitem{Bose3} D. Burgarth and S. Bose, Physical Review A \textbf{71},
507: 052315 (2005).
508:
509: \bibitem{Shitao1} T. Shi, B. Chen and Z. Song, Comm. Theor. Phys. \textbf{43}
510: 795 (2005).
511:
512: \bibitem{Shitao2} T. Shi, Ying Li, Z. Song and C. P. Sun, Phys. Rev. A
513: \textbf{71}, 032309 (2005).
514:
515: \bibitem{Fractional revival} R.W. Robinett, Physics Reports \textbf{392} 1
516: (2004)
517:
518: \bibitem{J-W} P. Jordan and E. Wigner, Z. Phys. \textbf{47}, 631 (1928); E.
519: Lieb, T. Schultz, and D. Mattis, Ann. Phys. \textbf{16}, 407 (1961); S.
520: Katsura, Phys. Rev. \textbf{127}, 1508 (1962).
521:
522: \bibitem{SS} F. Copper, A. Khare and U. Sukhatme. Phys. Rep., \textbf{251},
523: 267 (1995).
524: \end{thebibliography}
525:
526: \end{document}
527: