1: %% LyX 1.3 created this file. For more info, see http://www.lyx.org/.
2: %% Do not edit unless you really know what you are doing.
3: %\documentclass[twocolumn,english,prl,showpacs,amsmath,amssymb,superscriptaddress,floatfix]{revtex4}
4: %\usepackage{times}
5: %\usepackage[T1]{fontenc}
6: %\usepackage[latin1]{inputenc}
7: %\usepackage{amssymb}
8: \documentclass[twocolumn,amsmath,
9: amssymb,english,aps,prb,floatfix,showpacs]{revtex4}
10: \usepackage[english]{babel}
11: \usepackage{times}
12: \makeatletter
13:
14: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% LyX specific LaTeX commands.
15: %% Bold symbol macro for standard LaTeX users
16: %\newcommand{\boldsymbol}[1]{\mbox{\boldmath $#1$}}
17:
18:
19: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%% User specified LaTeX commands.
20: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% User specified LaTeX commands.
21: \usepackage{graphicx}
22:
23:
24: \usepackage{babel}
25: \makeatother
26: \begin{document}
27:
28: \title{Thermal entanglement of qubit pairs on the Shastry-Sutherland lattice}
29:
30:
31: \author{S. El Shawish}
32:
33:
34: \affiliation{J. Stefan Institute, Ljubljana, Slovenia}
35:
36:
37: \author{A. Ram\v{s}ak}
38:
39:
40: \affiliation{Faculty of Mathematics and Physics, University of Ljubljana, Ljubljana,
41: Slovenia}
42:
43:
44: \affiliation{J. Stefan Institute, Ljubljana, Slovenia}
45:
46:
47: \author{J. Bon\v{c}a}
48:
49:
50: \affiliation{Faculty of Mathematics and Physics, University of Ljubljana, Ljubljana,
51: Slovenia}
52:
53:
54: \affiliation{J. Stefan Institute, Ljubljana, Slovenia}
55:
56:
57: \date{2 April 2007}
58:
59: \begin{abstract}
60: We show that temperature and magnetic field properties of the
61: entanglement between spins on the two-dimensional Shastry-Sutherland
62: lattice can be qualitatively described by analytical results for a
63: qubit tetramer. Exact diagonalization of clusters with up to 20 sites
64: reveals that the regime of fully entangled neighboring pairs coincides
65: with the regime of finite spin gap in the spectrum. Additionally, the
66: results for the regime of vanishing spin gap are discussed and related
67: to the Heisenberg limit of the model.
68: \end{abstract}
69:
70: \pacs{75.10.Jm, 03.65.Yz, 03.67.Mn}
71:
72: \maketitle
73:
74: \section{Introduction}
75:
76: In any physical system with subsystems in interaction, individual
77: parts of the system are to some extent entangled, even if they are far
78: apart, as realized already at the beginning of modern quantum
79: mechanics sixty years ago. Today it has become appreciated that the
80: ability to establish entanglement between quantum particles in a
81: controlled manner is a crucial ingredient of any quantum information
82: processing system\cite{nielsen01}. On the other hand, it turned out
83: that the analysis of appropriately quantified entanglement between
84: parts of the system can also be a very useful tool in the study of
85: many body phenomena, as is, {\it e.g.}, the behavior of correlated
86: systems in the vicinity of crossovers between various regimes or even
87: points of quantum phase transition\cite{osterloh02}.
88:
89: Quantum entanglement of two distinguishable particles in a pure state
90: can be quantified through von Neuman entropy
91: \cite{bennett96,hill97,vedral97}. Entanglement between two
92: spin-${1 \over 2}$ particles -- qubit pair -- can be considered a
93: physical resource, an essential ingredient of algorithms suitable for
94: quantum computation. For a pair of subsystems A and B, each occupied
95: by a single electron, an appropriate entanglement measure is the
96: entanglement of formation, which can be quantified from the Wootters
97: formula\cite{wootters98}. In general, electron-qubits have the
98: potential for even richer variety of entanglement measure choices due
99: to both their charge and spin degrees of freedom. When entanglement is
100: quantified in systems of indistinguishable particles, the measure must
101: account for the effect of exchange and it must adequately deal with
102: multiple occupancy states
103: \cite{schliemann01,ghirardi04,eckert02,gittings02,zanardi02,vedral03}.
104: A typical example is the analysis of entanglement in lattice
105: fermion models (the Hubbard model, e.g.) where double occupancy
106: plays an essential role\cite{zanardi02}.
107:
108: In realistic hardware designed for quantum information processing,
109: several criteria for qubits must be fulfilled\cite{criteria}: the
110: existence of multiple identifiable qubits, the ability to initialize
111: and manipulate qubits, small decoherence, and the ability to measure
112: qubits, {\it i.e.}, to determine the outcome of computation. It seems that among
113: several proposals for experimental realizations of such quantum
114: information processing systems the criteria for scalable
115: qubits can be met in solid state structures consisting of coupled
116: quantum dots\cite{divincenzo05,coish06}. Due to the ability to
117: precisely control the number of electrons in such structures
118: \cite{elzerman03}, the entanglement has become experimentally
119: accessible quantity. In particular, recent experiments on
120: semiconductor double quantum dot devices have shown the evidence of
121: spin entangled states in GaAs based heterostuctures\cite{chen04} and
122: it was shown that vertical-lateral double quantum dots may be useful
123: for achieving two-electron spin entanglement\cite{hatano05}. It was
124: also demonstrated recently that in double quantum dot systems coherent
125: qubit manipulation and projective readout is possible \cite{petta05}.
126:
127: Qubit pairs to be used for quantum information processing must be to a
128: high degree isolated from their environment, otherwise small
129: decoherence requirement from the DiVincenzo's checklist can not be
130: fulfilled. The entanglement, {\it e.g.}, between two
131: antiferromagnetically coupled spins in contact with thermal bath, is
132: decreased at elevated temperatures and external magnetic
133: field\cite{nielsen00,arnesen01,wang01}, and will inevitably vanish at
134: some finite temperature\cite{fine05}.
135: %
136: Entanglement of a pair of electrons that are confined in a double
137: quantum dot is collapsed due to the Kondo effect at low
138: temperatures and for a very weak tunneling to the leads.
139: %
140: %Even at low temperatures and for a very weak tunneling to the
141: %leads will the entanglement of a pair of electrons confined in a
142: %double quantum dot be collapsed due to the Kondo effect.
143: %
144: At temperatures below the Kondo temperature a spin-singlet state
145: is formed between a confined electron and conduction electrons in
146: the leads\cite{rmzb06}. For other open systems there are many
147: possible sources of decoherence or phase-breaking, for example
148: coupling to phonon degrees of freedom \cite{yu02}.
149:
150: The main purpose of the present paper is to analyze the robustness of
151: the entanglement of spin qubit pairs in a planar lattice of spins
152: (qubits) with respect to frustration in magnetic couplings, elevated
153: temperatures as well as due to increasing external magnetic field. The
154: paper is organized as follows. Sec. \ref{secII} introduces the
155: model for two coupled qubit pairs -- qubit tetramer -- and presents
156: exact results for temperature and magnetic field dependence of the
157: entanglement between nearest and next-nearest-neighboring spins in a
158: tetrahedron topology. In Sec. \ref{secIII} the model is extended to
159: infinite lattice of qubit pairs described by the Shastry-Sutherland
160: model \cite{shastry81}. This model is convenient firstly, because of
161: the existence of stable spin-singlet pairs in the ground state in the
162: limit of weak coupling between the qubit pairs, and secondly, due to a
163: relatively good understanding of the physics of the model in the
164: thermodynamic limit. Entanglement properties of the Shastry-Sutherland
165: model were so far not considered quantitatively. Neverteless, several
166: results concerning the role of entanglement at a phase
167: transition in other low-dimensional spin lattice systems
168: \cite{osterloh02,syljuasen03,amico04,osborne02,roscilde04,roscilde05,larsson05},
169: as well as in fermionic systems\cite{gu04,deng05,legeza06} have
170: been reported recently. Near a quantum phase transition in some
171: cases entanglement even proves to be more efficient precursor of
172: the transition compared to standard spin-spin correlations
173: \cite{verstraete04,legeza06}. In Sec. \ref{secIV} we discuss
174: entanglement between nearest neighbors in the Heisenberg model,
175: representing a limiting case of the Shastry-Sutherland model. Results
176: are summarized in Sec. \ref{secV} and some technical details are given
177: in Appendix A.
178:
179: \section{Thermal entanglement of a qubit tetramer in magnetic field}
180: \label{secII}
181:
182: Consider first a double quantum dot composed of two adjacent quantum
183: dots weakly coupled via a controllable electron-hopping integral. By
184: adjusting a global back-gate voltage, precisely two electrons can be
185: confined to the dots. The inter-dot tunneling matrix element $t$
186: determines the effective antiferromagnetic (AFM) superexchange
187: interaction $J\sim 4t^{2}/U$, where $U$ is the scale of Coulomb
188: interaction between two electrons confined on the same dot. There are
189: several possible configurations of coupling between such double
190: quantum dots. One of the simplest specific designs is shown
191: schematically in Fig.~\ref{fig1}(a): four qubits at vertices of a
192: tetrahedron. In addition to the coupling A-B, by appropriate
193: arrangements of gate electrodes the tunneling between A-C and A-D can
194: as well be switched on.
195: %
196: \begin{figure}
197: \begin{center}
198: \includegraphics[width=60mm,keepaspectratio]{Fig1.eps}
199: \end{center}
200:
201: \caption{(Color online) (a) Two coupled qubit pairs (dimers) in
202: tetrahedral topology. (b) Shastry-Sutherland lattice as realized, {\it e.g.},
203: in the SrCu$_{2}$(BO$_{3}$)$_{2}$ compound.}
204:
205: \label{fig1}
206: \end{figure}
207:
208: We consider here the case where $J/U \ll 1$, thus double occupancy of
209: individual dot is negligible and appropriate Hilbert space is spanned
210: by two dimers (qubit pairs): spins at
211: sites A-B and C-D
212: are coupled by effective AFM Heisenberg magnetic exchange $J$ and at sites
213: A-C, B-C, A-D, B-D by $J'$. The corresponding
214: hamiltonian of such a pair of dimers is given as
215: %
216: \begin{eqnarray}
217: \label{habcd}
218: H_4&=&J({\textbf{S}}_{\textrm{A}}\cdot{\textbf{S}}_{\textrm{B}}+
219: {\textbf{S}}_{\textrm{C}}\cdot{\textbf{S}}_{\textrm{D}})+\\ \nonumber
220: &+&2J^\prime({\textbf{S}}_{\textrm{A}}\cdot{\textbf{S}}_{\textrm{C}}+
221: {\textbf{S}}_{\textrm{B}}\cdot{\textbf{S}}_{\textrm{C}}+
222: {\textbf{S}}_{\textrm{A}}\cdot{\textbf{S}}_{\textrm{D}}+
223: {\textbf{S}}_{\textrm{B}}\cdot{\textbf{S}}_{\textrm{D}})- \\ \nonumber
224: &-&B(S^z_{\textrm{A}}+S^z_{\textrm{B}}+S^z_{\textrm{C}}+S^z_{\textrm{D}}), \nonumber
225: \end{eqnarray}
226: where $\textbf{S}_i={1 \over 2} {\bf \sigma}_i$ is spin operator
227: corresponding to the site $i$ and $B$ is external homogeneous magnetic
228: field in the direction of the $z$-axis. Factor 2 in Eq.~(\ref{habcd})
229: is introduced for convenience -- such a parameterization represents
230: the simplest case of finite Shastry-Sutherland lattice with periodic
231: boundary conditions studied in Sec. \ref{secIII}.
232:
233:
234: %
235: \begin{figure*}
236: \begin{center}
237: \includegraphics[width=60mm,keepaspectratio]{Fig2a.eps}
238: \includegraphics[width=60mm,keepaspectratio]{Fig2b.eps}\\
239: \includegraphics[width=60mm,keepaspectratio]{Fig2c.eps}
240: \includegraphics[width=60mm,keepaspectratio]{Fig2d.eps}
241: \end{center}
242:
243: \caption{(Color online) (a) Zero-temperature concurrence
244: $C_{\mathrm{A} \mathrm{B}}$ as a function of $J'/J$ and $B/J$.
245: Different regimes are characterized by particular ground state
246: functions $ |\phi_n\rangle$ defined in Appendix A. (b) $T/J=0.1$
247: results for $C_{\mathrm{A} \mathrm{B} }$. (c) Next nearest concurrence
248: $C_{\mathrm{A} \mathrm{C} }$ for $T=0$, and (d) for $T/J=0.1$.
249: Dashed lines separate $C_{\mathrm{A}\mathrm{B}(\mathrm{C})}>0$ from
250: $C_{\mathrm{A}\mathrm{B}(\mathrm{C})}=0$.}
251:
252: \label{fig2}
253: \end{figure*}
254:
255: \subsection{Concurrence}
256:
257: We focus here on the entanglement properties of two coupled qubit
258: dimers. The entanglement of a pair of spin qubits A and B may be
259: defined through concurrence\cite{bennett96}, $C =
260: 2|\alpha_{\uparrow\!\uparrow}\alpha_{\downarrow\!\downarrow}-
261: \alpha_{\uparrow\!\downarrow}\alpha_{\downarrow\!\uparrow}|$, if the
262: system is in a pure state
263: $|\Psi_{\mathrm{AB}}\rangle=\sum_{ss'}\alpha_{ss'}|s\rangle_{\!
264: \mathrm{A}} |s'\rangle_{\! \mathrm{B}}$, where $|s \rangle_i$
265: corresponds to the basis $|\!\uparrow\,\rangle_i$,
266: $|\!\downarrow\,\rangle_i$. Concurrence varies from $C=0$ for an
267: unentangled state (for example $|\!\uparrow\,\rangle_\mathrm{A}
268: |\!\uparrow\,\rangle_\mathrm{B}$) to $C=1$ for completely entangled
269: Bell states\cite{bennett96} $ {1 \over \sqrt{2}}(
270: |\!\uparrow\,\rangle_\mathrm{A} |\!\uparrow\,\rangle_\mathrm{B} \pm
271: |\!\downarrow\,\rangle_\mathrm{A} |\!\downarrow\,\rangle_\mathrm{B})$
272: or ${1 \over \sqrt{2}}(|\!\uparrow\,\rangle_\mathrm{A}
273: |\!\downarrow\,\rangle_\mathrm{B} \pm
274: |\!\downarrow\,\rangle_\mathrm{A} |\!\uparrow\,\rangle_\mathrm{B})$.
275:
276: For finite inter-pair coupling $J^\prime \ne 0$ or at elevated
277: temperatures the A-B pair can not be described by a pure state. In the
278: case of mixed states describing the subsystem A-B the concurrence may
279: be calculated from the reduced density matrix
280: $\rho_{\textrm{A}\textrm{B}}$ given in the standard basis $|s
281: \rangle_i |s' \rangle_j$\cite{wootters98}. Concurrence can be further
282: expressed in terms of spin-spin correlation functions
283: \cite{osterloh02,syljuasen03}, where for systems that are axially
284: symmetric in the spin space the concurrence may conveniently be given in a
285: simple closed form\cite{ramsak06}, which for the thermal
286: equilibrium case simplifies further,
287: %
288: \begin{equation}
289: C_{\mathrm{A} \mathrm{B} } =2 \textrm{max}(0,
290: |\langle S_{\mathrm{A}}^{+}
291: S_{\mathrm{B}}^{-}\rangle|-\sqrt{\langle P_{\mathrm{A}}^
292: {\uparrow}P_{\mathrm{B}}^{\uparrow}\rangle\langle P_{\mathrm{A}}^
293: {\downarrow}P_{\mathrm{B}}^{\downarrow}\rangle}).
294: \label{cmax}
295: \end{equation}
296: %
297: Here $S_i^{+} = (S_i^{-})^{\dagger} = S_i^x+\imath S_i^y$ is the spin
298: raising operator for dot $i$ and $P_i^\uparrow={1 \over 2}(1+2
299: S_i^z)$, $P_i^\downarrow={1 \over 2}(1-2 S_i^z)$ are the projection
300: operators onto the state $|\!\uparrow\,\rangle_i$ or
301: $|\!\downarrow\,\rangle_i$, respectively. We consider the concurrence
302: at fixed temperature, therefore the expectation values in the
303: concurrence formula Eq.~(\ref{cmax}) are evaluated as
304: %
305: \begin{equation}
306: \langle {\cal O} \rangle = {1 \over Z} \sum_n \langle n| {\cal O}|n \rangle e^{-\beta E_n},
307: \label{o}
308: \end{equation}
309: %
310: where $Z= \sum_n e^{-\beta E_n}$ is the partition function,
311: $\beta=1/T$, and $\{|n \rangle \}$ is a complete set of states of the
312: system. Note that due to the equilibrium and symmetries of the system,
313: several spin-spin correlation functions vanish, $\langle
314: S_{\mathrm{A}}^{+}S_{\mathrm{B}}^{+}\rangle=0$, for example.
315:
316: In vanishing magnetic field, where the SU(2) symmetry is restored, the
317: concurrence formula Eq.~(\ref{cmax}) simplifies further and is
318: completely determined by only one\cite{werner89} spin invariant
319: $\langle\textrm{\bf S}_{\mathrm{A}}\cdot\textrm{\bf
320: S}_{\mathrm{B}}\rangle$,
321: %
322: \begin{equation}
323: C_{\mathrm{A} \mathrm{B} } = \textrm{max}(0,-2\langle\textrm{\bf
324: S}_{\mathrm{A}}\cdot\textrm{\bf
325: S}_{\mathrm{B}}\rangle-\frac{1}{2}).
326: \label{su2}
327: \end{equation}
328: %
329: The concurrence may be expected to be significant whenever enhanced
330: spin-spin correlations indicate A-B singlet formation.
331:
332:
333: \subsection{Analytical results}
334:
335: There are several known results related to the model
336: Eq.~(\ref{habcd}). In the special case of $J^\prime=0$, for example,
337: the tetramer consists of two decoupled spin dimers with concurrence
338: $C_{\mathrm{A} \mathrm{B}}$ (or the corresponding thermal
339: entanglement) as derived in Refs.~\onlinecite{nielsen00,arnesen01}.
340: Entanglement of a qubit pair described by the related XXZ Heisenberg
341: model with Dzyaloshinskii-Moriya anisotropic interaction can be also
342: obtained analytically\cite{wang01}. Hamiltonian $H_4$ with additional
343: four-spin exchange interaction but in the absence of magnetic field
344: was considered recently in the various limiting cases\cite{bose05}.
345:
346: Tetramer model Eq.~(\ref{habcd}) considered here is exactly solvable
347: and in Appendix A we present the corresponding eigenvectors and
348: eigenenergies. The concurrence $C_{\mathrm{A}
349: \mathrm{B}}$ is for this case determined from Eq.~(\ref{cmax}) with
350: %
351: \begin{eqnarray}
352: \langle S_{\rm A}^+S_{\rm B}^-\rangle&=&\frac{1}{Z}\Big[\!\!
353: -e^{3j/2}/2-e^{j/2}\left(e^{b}+e^{-b}\right)/2
354: \nonumber\\
355: &+&e^{-j/2+4j'}/6
356: +e^{-j/2+2j'}\left(e^{b}+e^{-b}\right)/4
357: \nonumber\\
358: &+&e^{-j/2-2j'}\left(e^{b}/4+1/3
359: +e^{-b}/4\right)\!\!\Big],
360: \end{eqnarray}
361: %
362: \noindent
363: where $j=\beta J$, $j'=\beta J'$, $b=\beta B$, and with
364: %
365: \begin{eqnarray}
366: \langle P_{\rm A}^{\uparrow\downarrow}
367: P_{\rm B}^{\uparrow\downarrow}\rangle&=&\frac{1}{Z}\Big[
368: e^{j/2}\left(e^{\pm b}\right)+e^{-j/2+4j'}/3
369: \nonumber\\
370: &+&e^{-j/2+2j'}\left(1+e^{\pm b}\right)/2
371: \nonumber\\
372: &+&e^{-j/2-2j'}\left(1/6+e^{\pm b}/2+e^{\pm2 b}
373: \right)\!\!\Big].
374: \end{eqnarray}
375: %
376: Here
377: %
378: \begin{eqnarray}
379: Z&=&e^{3j/2}
380: +2e^{j/2}\left(e^{b}+1+e^{-b}\right)
381: \nonumber\\
382: &+&e^{-j/2+4j'}
383: +e^{-j/2+2j'}\left(e^{b}+1+e^{-b}\right)
384: \nonumber\\
385: &+&e^{-j/2-2j'}\left(e^{2b}+e^{b}+1
386: +e^{-b}+e^{-2b}\right)
387: \end{eqnarray}
388: %
389: is the partition function.
390:
391:
392: %
393: \begin{figure}
394: \begin{center}
395: \includegraphics[width=60mm,keepaspectratio]{Fig3a.eps}\\
396: \includegraphics[width=60mm,keepaspectratio]{Fig3b.eps}
397: \end{center}
398:
399: \caption{(Color online) (a) Temperature and magnetic field dependence of
400: $C_{\mathrm{A}\mathrm{B}}$ for $J'/J=0.4$ and (b) $J'=J$.
401: Dashed lines separate $C_{\mathrm{A}\mathrm{B}}>0$ from
402: $C_{\mathrm{A}\mathrm{B}}=0$.
403: }
404:
405: \label{fig3}
406: \end{figure}
407:
408:
409: Alternatively, one can define and analyze also the entanglement
410: between spins at sites A and C and the corresponding concurrence
411: $C_{\mathrm{A} \mathrm{C}}$ can be expressed from Eq.~(\ref{cmax}) by
412: applying additional correlators with replaced B$\to$C,
413: %
414: \begin{eqnarray}
415: \langle S_{\rm A}^+S_{\rm C}^-\rangle&=&\frac{1}{Z}\Big[\!\!
416: -e^{-j/2+4j'}/3
417: \nonumber\\
418: &-&e^{-j/2+2j'}\left(e^{b}+e^{-b}\right)/4
419: \nonumber\\
420: &+&e^{-j/2-2j'}\left(e^{b}/4+1/3
421: +e^{-b}/4\right)\!\!\Big],
422: \end{eqnarray}
423: %
424: and
425: %
426: \begin{eqnarray}
427: \langle P_{\rm A}^{\uparrow\downarrow}
428: P_{\rm C}^{\uparrow\downarrow}\rangle&=&\frac{1}{Z}\Big[
429: e^{3j/2}/4+e^{j/2}\left(1/2+e^{\pm b}\right)
430: \nonumber\\
431: &+&e^{-j/2+4j'}/12
432: +e^{-j/2+2j'} e^{\pm b}/2
433: \nonumber\\
434: &+&e^{-j/2-2j'}\left(1/6
435: +e^{\pm b}/2+e^{\pm2 b}\right)\!\!\Big].
436: \end{eqnarray}
437:
438: The line $2J'=J$ represents a particularly interesting special case
439: where two dimers are
440: coupled symmetrically forming a regular tetrahedron. An important
441: property of this system is the (geometrical) frustration of, {\it
442: e.g.}, qubits C-A-B. Such a frustration is the driving force of the
443: quantum phase transition found in the Shastry-Sutherland model and is
444: the reason for similarity of the results for two coupled dimers and a
445: large planar lattice studied in the next Section.
446:
447:
448: \subsection{Examples}
449:
450: In the low temperature limit the concurrence is determined by the
451: ground state properties while transitions between various regimes are
452: determined solely by crossings of eigenenergies, which depend on two
453: parameters $(J^\prime/J,B/J)$. There are 5 distinct regimes for
454: $C_{\mathrm{A} \mathrm{B}}$ shown in Fig.~\ref{fig2}(a): (i)
455: completely entangled dimers (singlets A-B and C-D, state
456: $|\phi_1\rangle$ from Appendix A), $C_{\mathrm{A}
457: \mathrm{B}}=1$; (ii) for $B>J$ and smaller $J^\prime/J$ the
458: concurrence is zero because the energy of the state consisted of a
459: product of fully polarized A-B and C-D triplets, $|
460: \phi_{12}\rangle$, is the lowest energy in this regime; (iii)
461: concurrence is zero also for $J^\prime>J/2$ and low $B/J$, with the
462: ground state $| \phi_2 \rangle $. There are two regimes corresponding
463: to ${1 \over 2}$ step in $C_{\mathrm{A} \mathrm{B}}$ where the ground
464: state is either (iv) any linear combination of degenerate states $|
465: \phi_{6,7}\rangle$, {\it i.e.}, simultaneous A-B singlet (triplet) and
466: C-D triplet (singlet) for $J^\prime<J/2$, or (v) state $|
467: \phi_5\rangle$ at $J^\prime>J/2$ and larger $B$. Qubits A-C are due to
468: special topology never fully entangled, and the corresponding
469: $C_{\mathrm{A} \mathrm{C}}$ is presented in Fig.~\ref{fig2}(c). In the
470: limit of $J^\prime \gg J$ the tetramer corresponds to a Heisenberg
471: model ring consisted of 4 spins and in this case qubit A is due to
472: tetramer symmetry equally entangled to both neighbors (C and D), thus
473: $C_{\mathrm{A} \mathrm{C}}={1 \over 2}$.
474:
475:
476: At elevated temperatures the concurrence is smeared out as shown in
477: Figs.~\ref{fig2}(b,d). Note the dip separating the two different
478: regimes with $C_{\mathrm{A} \mathrm{B}}={1 \over 2}$, seen also in the
479: $C_{\mathrm{A} \mathrm{C}}={1 \over 2}$ case. This dip clearly
480: separates different regimes discussed in the previous $T=0$ limit and
481: signals a proximity of a disentangled excited state. For sufficiently
482: high temperatures vanishing concurrence is expected
483: \cite{fine05}. The critical temperature $T_c$ denoted by a dashed line
484: is set by the magnetic exchange scale $J$, since
485: at higher temperatures local singlets are broken irrespectively of
486: the magnetic field.
487:
488: A rather unexpected result is shown in Fig.~\ref{fig3}(a) where at
489: $B\gtrsim 2 J$ and low temperatures the concurrence slightly increases
490: with increasing temperature due to the contribution of excited A-B
491: singlet components that are absent in the ground state. Similar
492: behavior is found for $J^\prime\sim0$ around $B\sim J$, which is
493: equivalent to the case of a single qubit
494: dimer\cite{nielsen00,arnesen01} (not shown here). There is no
495: distinctive feature in temperature and magnetic field dependence of
496: $C_{\mathrm{A} \mathrm{B}}$ when $J'>J/2$ and a typical results is
497: shown in Fig.~\ref{fig3}(b) for $J'>J$.
498:
499:
500:
501:
502: \section{Planar array of qubit pairs: the Shastry-Sutherland lattice}
503: \label{secIII}
504: \subsection{Preliminaries}
505:
506: The central point of this paper is the analysis of pair entanglement
507: for the case of a larger number of coupled qubit pairs. In the
508: following it will be shown that the results corresponding to tetramers
509: considered in the previous Section can be very helpful for better
510: understanding pair-entanglement of $N>4$ qubits. There are several
511: possible generalizations of coupled dimers and one of the simplest in
512: two dimensions is the Shastry-Sutherland lattice shown in
513: Fig.~\ref{fig1}(b). Neighboring sites A-B are connected with exchange
514: interaction $J$ and next-neighbors with $J'$. The corresponding
515: hamiltonian for $N/2$ dimers ($N$ sites) is given with
516: %
517: \begin{equation}
518: H_N=J\sum_{\{\textrm{AB\}}}{\textbf{S}}_{i}\cdot{\textbf{S}}_{j}+
519: J'\sum_{\{\textrm{AC\}}}{\textbf{S}}_{i}\cdot{\textbf{S}}_{j}-B\sum_{i=1}^{N}S_{i}^{z}.
520: \end{equation}
521: %
522: Periodic boundary conditions are used. For the special case $N=4$ the
523: model reduces to Eq.~(\ref{habcd}) where due to periodic boundary
524: conditions sites A-C (and other equivalent pairs) are doubly
525: connected, therefore a factor of 2 in Eq.~(\ref{habcd}), as mentioned
526: in Sec. \ref{secII}.
527:
528: %---------------------------------------------------------------
529: % new - start
530: %---------------------------------------------------------------
531: % toy model, povezava eksperiment
532: The Shastry-Sutherland model (SSM) was initially proposed as a toy
533: model possessing an exact dimerized eigenstate known as a valence bond
534: crystal \cite{shastry81}. Recently, the model has experienced a sudden
535: revival of interest by the discovery of the two-dimensional
536: spin-liquid compound SrCu$_{2}$(BO$_{3}$)$_{2}$
537: \cite{smith91,kageyama99} since it is believed that magnetic properties
538: of this compound are reasonably well described by the SSM
539: % for particular values of the
540: %AFM couplings
541: \cite{miyahara03}. In fact, several generalizations of the SSM
542: have been introduced to account better for recent high-resolution
543: measurements revealing the magnetic fine structure of
544: SrCu$_{2}$(BO$_{3}$)$_{2}$
545: \cite{miyahara03,jorge03,elshawish1,elshawish2}. Soon after the
546: discovery of the SrCu$_{2}$(BO$_{3}$)$_{2}$ system, the SSM thus
547: became a focal point of theoretical investigations in the field of
548: frustrated AFM spin systems, particularly low-dimensional quantum
549: spin systems where quantum fluctuations lead to magnetically
550: disordered ground states (spin liquids) with a spin gap in the
551: excitation spectrum.
552:
553: % variation of J'/J
554: The SSM is a two-dimensional frustrated antiferromagnet with a unique
555: spin-rotation invariant exchange topology that leads in the limit
556: $J\gg J'$ to an exact gapped dimerized ground state with localized
557: spin singlets on the dimer bonds (dimer phase). In the opposite limit,
558: $J\ll J'$, the model becomes ordinary AFM Heisenberg model with a
559: long-range N\' eel order and a gapless spectrum (N\' eel phase). While
560: two of the phases are known, there are still open questions regarding
561: the existence and the nature of the intermediate phases. Several possible
562: scenarios have been proposed, {\it e.g.}: either a direct transition between the
563: two states occurs at the quantum critical point near $J'/J\sim 0.7$
564: \cite{miyahara99,mueller00}, or a transition via an intermediate phase
565: that exists somewhere in the range of $J'/J>0.6$ and $J'/J<0.9$
566: \cite{isacsson06}. Although different theoretical approaches have been
567: applied, a true nature of the intermediate phase (if any) has still
568: not been settled. As will be evident later on, our
569: exact-diagonalization results support the first scenario.
570:
571: % variation of B
572: The SSM phase diagram reveals interesting behavior also for varying
573: external magnetic field. In particular, experiments on
574: SrCu$_{2}$(BO$_{3}$)$_{2}$ in strong magnetic fields show formation of
575: magnetization plateaus \cite{kageyama99,onizuka00}, which are believed
576: to be a consequence of repulsive interaction between almost localized
577: spin triplets. Several theoretical approaches support the idea that
578: most of these plateaus are readily explained within the (bare) SSM
579: \cite{miyahara99,momoi00,misguich01}. Recent variational treatment
580: based on entangled spin pairs revealed new insight into various phases
581: of the SSM\cite{isacsson06}.
582:
583: %finite T
584: Although extensively studied, the zero-temperature phase diagram
585: of the SSM remains elusive. This lack of reliable solutions is
586: even more pronounced when considering thermal fluctuations in SSM
587: as only few methods allow for the inclusion of finite temperatures
588: in frustrated spin systems. In this respect, the calculation of
589: thermal entanglement between the spin pairs would also provide a
590: new insight into the complexity of the SSM.
591:
592: %
593: \begin{figure}
594: \begin{center}
595: \includegraphics[width=70mm,keepaspectratio]{Fig4.eps}
596: \end{center}
597:
598: \caption{(Color online) Results for the Shastry-Sutherland lattice with
599: $N=20$ sites and periodic boundary conditions. Presented are
600: renormalized spin-spin correlation functions $-2\langle\textrm{\bf
601: S}_{\mathrm{A}}\cdot\textrm{\bf S}_{\mathrm{B,C}}\rangle-\frac{1}{2}$
602: as a function of $J'/J$ and for various temperatures. Asterisk
603: indicates critical $J'_c$ which roughly separates the dimer and N\'
604: eel phase.}
605:
606: \label{fig4}
607: \end{figure}
608:
609:
610: %
611: \begin{figure*}
612: \begin{center}
613: \includegraphics[width=60mm,keepaspectratio]{Fig5a.eps}
614: \includegraphics[width=60mm,keepaspectratio]{Fig5b.eps}\\
615: \includegraphics[width=60mm,keepaspectratio]{Fig5c.eps}
616: \includegraphics[width=60mm,keepaspectratio]{Fig5d.eps}
617: \end{center}
618:
619: \caption{(Color online) (a) Zero-temperature concurrence
620: $C_{\mathrm{A} \mathrm{B} }$ for a 20-site cluster for various $J'/J$
621: and $B/J$. Shaded area represents the regime of fully entangled
622: dimers, $C_{\mathrm{A} \mathrm{B} }=1$. (b) The corresponding results
623: for $T/J=0.1$. (c) Next nearest concurrence $C_{\mathrm{A} \mathrm{C}
624: }$ for $T=0$, and (d) for $T/J=0.1$. Note qualitative and even
625: quantitative similarity with the tetramer results, Fig.~\ref{fig2}.
626: Dashed lines separate $C_{\mathrm{A}\mathrm{B}(\mathrm{C})}>0$ from
627: $C_{\mathrm{A}\mathrm{B}(\mathrm{C})}=0$.
628: }
629: \label{fig5}
630: \end{figure*}
631:
632:
633: \subsection{Numerical method}
634: %---------------------------------------------------------------
635: % new - start
636: %---------------------------------------------------------------
637: We use the low-temperature Lanczos method \cite{ltlm} (LTLM), an
638: extension of the finite-temperature Lanczos method \cite{jj1} (FTLM)
639: for the calculation of static correlation functions at low
640: temperatures. Both methods are nonperturbative, based on the Lanczos
641: procedure of exact diagonalization and random sampling over different
642: initial wave functions. A main advantage of LTLM is that it accurately
643: connects zero- and finite-temperature regimes with rather small
644: numerical effort in comparison to FTLM. On the other hand, while FTLM
645: is limited in reaching arbitrary low temperatures on finite systems,
646: it proves to be computationally more efficient at higher
647: temperatures. A combination of both methods therefore provides
648: reliable results in a wide temperature regime with moderate
649: computational effort. We note that FTLM was in the past successfully
650: used in obtaining thermodynamic as well as dynamic properties of
651: different models with correlated electrons as are: the $t$-$J$
652: model,\cite{jj1} the Hubbard model,\cite{bonca02} as well as the SSM
653: model.\cite{jorge03,elshawish2}
654:
655: In comparison with the conventional Quantum Monte Carlo (QMC) methods
656: LTLM possesses the following advantages: (i) it does not suffer from
657: the minus-sign problem that usually hampers QMC calculations of
658: many-electron as well as frustrated spin systems, (ii) the method
659: continuously connects the zero- and finite-temperature regimes, (iii)
660: it incorporates as well as takes the advantage of the symmetries of
661: the problem, and (iv) it yields results of dynamic properties in the
662: real time in contrast to QMC calculations where imaginary-time Green's
663: function is obtained. The LTLM (FTLM) is on the other hand limited to
664: small lattices which usually leads to sizable finite-size effects. To
665: account for these, we applied LTLM to different square lattices with
666: $N=8,16$ and 20 sites using periodic boundary conditions (we note that
667: next-larger system, $N=32$, was too large to be handled
668: numerically). Another drawback of the LTLM (FTLM) is the difficulty of
669: the Lanczos procedure to resolve degenerate eigenstates that emerge
670: also in the SSM.
671: %
672: %(Since the unit cell of the SS lattice contains 2 dimers, the lowest,
673: %single-triplet excitation above the dimerized ground state becomes
674: %degenerate even at finite magnetic fields \cite{}.)
675: %
676: In practice, this manifests itself in severe statistical fluctuations
677: of the calculated amplitude for $T\to 0$ since in this regime only a
678: few (degenerate) eigenstates contribute to thermal average. The
679: simplest way to overcome this is to take a larger number of random
680: samples $R\gg 1$, which, however, requires a longer CPU time. We have,
681: in this regard, also included a small portion of anisotropy in the SSM
682: (in the form of the anisotropic interdimer Dzyaloshinskii-Moriya interaction $D^z
683: \sum_{\{\rm{AC}\}} (S_A^xS_C^y-S_A^yS_C^x)$, $D^z/J\sim 0.01$),
684: which slightly splits the doubly degenerate single-triplet levels.
685: In this way, $R\sim 30$ per $S^z$ sector was enough for all
686: calculated curves to converge within $\sim 1\%$ for $T/J<1$. Here,
687: the number of Lanczos iterations $M=100$ was used along with the
688: full reorthogonalization of Lanczos vectors at each step.
689: %---------------------------------------------------------------
690: % new - end
691: %---------------------------------------------------------------
692:
693: \subsection{Entanglement}
694:
695: Entanglement in the absence of magnetic field is most prominently
696: reflected in spin-spin correlation functions, {\it e.g.},
697: $\langle\textrm{\bf S}_{\mathrm{A}}\cdot \textrm{\bf
698: S}_{\mathrm{B}}\rangle $ and $\langle\textrm{\bf S}_{\mathrm{A}}\cdot
699: \textrm{\bf S}_{\mathrm{C}}\rangle $. In zero temperature limit due to
700: quantum phase transition at $J_c'$ these correlations change sign. In
701: Fig.~\ref{fig4} are presented renormalized spin-spin correlation
702: functions (for positive values identical to concurrence) as a function
703: of $J'/J$: (i) $C_{\mathrm{A} \mathrm{B} }>0$ in dimer phase and (ii)
704: $C_{\mathrm{A} \mathrm{C} }>0$ in the N{\' e}el phase. Critical $J'_c$
705: is indicated by asterisk. The results for $N=16$ are qualitatively and
706: quantitatively similar to the $N=20$ case presented here. At finite
707: temperatures spin correlations are smeared out as shown in
708: Fig.~\ref{fig4} for various $T$. Limiting Heisenberg case, $J'\to
709: \infty$, is discussed in more detail in the next Section. $J'=0$ case
710: corresponds to the single dimer limit\cite{arnesen01} and Sec.
711: \ref{secII}.
712:
713: Complete phase diagram of the SSM at $T=0$ but with finite magnetic
714: field can be classified in terms of concurrence instead of spin
715: correlations. In Fig.~{\ref{fig5}(a)$C_{\mathrm{A}
716: \mathrm{B}}$ is presented as a function of $(J'/J,B/J)$ as in the case of a single
717: tetramer, Fig.~{\ref{fig2}(a). Presented results correspond to the
718: $N=20$ case, while $N=16$ system exhibits very similar structure (not
719: shown here). $N=8$ and $N=4$ cases are qualitatively similar, the main
720: difference being the value of critical $J'_c$ which increases with
721: $N$. Remarkable similarity between all these cases can be interpreted
722: by local physics in the regime of finite spin gap, $J'<J'_c$. Qubit
723: pairs are there completely entangled, $C_{\mathrm{A} \mathrm{B}}=1$,
724: and $C_{\mathrm{A} \mathrm{B}}\sim {1\over 2}$ for magnetic field
725: larger than the spin gap, but $B<J+2 J'$. For even larger $B$
726: concurrence approaches zero, similar to the $N=4$ case. Concurrence is
727: zero also for $J'>J'_c$, except along the $B\sim 4 J'$ line where weak
728: finite concurrence could be the finite size effect. Similar results
729: are found also for $N=16,8$ cases, and are most pronounced in the
730: $N=4$ case. At finite temperature the structure of concurrence is
731: smeared out [Fig.~\ref{fig5}(b)] similar to Fig.~\ref{fig2}(b).
732:
733: Concurrence $C_{\mathrm{A} \mathrm{C}}$ corresponding to next-nearest
734: neighbors is, complementary to $C_{\mathrm{A} \mathrm{B}}$, increased
735: in the N\' eel phase of the diagram, Fig.~\ref{fig5}(c). The
736: similarity with $N=4$, Fig.~\ref{fig2}(c) is somewhat surprising
737: because in this regime long-range correlations corresponding to the
738: gapless spectrum of AFM-like physics are expected to change also short
739: range correlations. The only quantitative difference compared to $N=4$
740: is the maximum value of $C_{\mathrm{A} \mathrm{C}}\sim 0.3$ instead of
741: $0.5$ (beside the critical value $J'_c$ discussed in the previous
742: paragraph). Concurrence is very small for $B>J+2 J'$. At finite
743: temperatures fine fluctuations in the concurrence structure are
744: smeared out, Fig.~\ref{fig5}(d).
745:
746: Temperature and magnetic field dependence of $C_{\mathrm{A}
747: \mathrm{B}}$ in the dimer phase is presented in Fig.~\ref{fig6}
748: for fixed $J'/J=0.4$. Similarity with the corresponding $N=4$
749: tetramer case, Fig.~\ref{fig3}(a), is astonishing and is again the
750: consequence of local physics in the presence of a finite spin gap.
751: Finite size effects (in comparison with $N=16$ and $N=8$ cases)
752: are very small (not shown). Dashed line represents the borderline
753: of the $C_{\mathrm{A} \mathrm{B}}=0$ region: critical $T_c\approx
754: 0.75 J$ valid for $B/J\lesssim 3$, that is in this regime nearly
755: independent of $B$, is slightly larger than in the single tetramer
756: case where its insensitivity to $B$ is even more pronounced.
757:
758: %
759: \begin{figure}
760: \begin{center}
761: \includegraphics[width=60mm,keepaspectratio]{Fig6.eps}
762: \end{center}
763:
764: \caption{(Color online) Temperature and magnetic field dependence of
765: $C_{\mathrm{A}\mathrm{B}}$ for $J'/J=0.4$ and $N=20$. Note the
766: similarity with the corresponding tetramer results,
767: Fig.~\ref{fig3}(a). Dashed lines separate $C_{\mathrm{A}\mathrm{B}}>0$ from
768: $C_{\mathrm{A}\mathrm{B}}=0$.}
769:
770: \label{fig6}
771: \end{figure}
772:
773: \section{Heisenberg limit}
774: \label{secIV}
775:
776: The concurrence corresponding to next-nearest neighbors in SSM,
777: $C_{\mathrm{A}\mathrm{C}}$, is non zero in the N\' eel phase for
778: $J'>J'_c$. Typical result for concurrence in this regime (for
779: fixed $J'/J=1$) in terms of temperature and magnetic field is
780: presented in Fig.~\ref{fig7}(a). At zero temperature the
781: concurrence is zero for $B>4J'$ [compare with Fig.~\ref{fig2}(c)
782: and Fig.~\ref{fig5}(c)].
783:
784: In the limit $J=0$ the model simplifies to the AFM Heisenberg model on
785: a square lattice of $N$ sites,
786: %
787: \begin{equation}
788: H_{\rm AC}=J' \sum_{\{\rm AC\}}{\bf S}_i\cdot {\bf S}_j-B\sum_{i=1}^{N} S_i^z.
789: \label{heisenberg}
790: \end{equation}
791: %
792: Several results for this model have already been presented for very
793: small clusters\cite{wang012,cao05,zhang05}, however the temperature
794: and magnetic field dependence of the concurrence for systems with
795: sufficiently large number of states and approaching thermodynamic
796: limit has not been presented so far.
797:
798:
799: %
800: \begin{figure}
801: \begin{center}
802: \includegraphics[width=60mm,keepaspectratio]{Fig7a.eps}\\
803: \includegraphics[width=60mm,keepaspectratio]{Fig7b.eps}
804: \end{center}
805: \caption{(Color online) (a) Next nearest neighbor concurrence
806: $C_{\mathrm{A}\mathrm{C}}$ for $J'=J$. (b) Heisenberg lattice result
807: as a special case of the SSM, $J=0$. Shaded region represents
808: $C_{\mathrm{A}\mathrm{C}}=0$. In the line shaded region (low
809: finite temperature and large magnetic field) our numerical results
810: set only the upper limit
811: $C_{\mathrm{A}\mathrm{C}}<5\cdot 10^{-4}$.
812: Dashed lines separate $C_{\mathrm{A}\mathrm{C}}>0$ from
813: $C_{\mathrm{A}\mathrm{C}}=0$.}
814:
815: \label{fig7}
816: \end{figure}
817:
818: %
819: \begin{figure}
820: \begin{center}
821: \includegraphics[width=75mm,keepaspectratio]{Fig8.eps}
822: \end{center}
823:
824: \caption{(Color online) (a) Zero-temperature concurrence
825: $C_{\mathrm{A}\mathrm{C}}$ in the Heisenberg limit as a function of
826: $B/J'$ and for various $N=4$, $8$, $16$, $20$. Sections with
827: different total spin values are additionally labeled. (b) Finite-size
828: scaling of concurrence in the absence of magnetic field. Full line
829: represents the fit corresponding to Ref.~\onlinecite{manousakis},
830: $C_{\mathrm{A}\mathrm{C}}\approx {1\over 6}+2N^{-3/2}$. }
831:
832: \label{fig8}
833: \end{figure}
834:
835:
836: In Fig.~\ref{fig7}(b) we further presented temperature and
837: magnetic field dependence of concurrence for the Heisenberg model
838: for $N=20$ (results for $N=16$ are quantitatively similar, but not
839: shown here). Temperature and magnetic field dependence of
840: $C_{\mathrm{A}\mathrm{C}}$ exhibits peculiar semi-island shape
841: where at fixed value of $B$ the concurrence increases with
842: increasing temperature. This effect is to some extent seen in all
843: cases and is the consequence of exciting local singlet states,
844: which do not appear in the ground state. At $T\to 0$ finite steps
845: with increasing $B$ correspond to gradual transition from the
846: singlet ground state to totally polarized state with total spin
847: $S=10$ and vanishing concurrence. This is in more detail presented
848: in Fig.~\ref{fig8}(a) for various $N=4, 8, 16, 20$. At $B=0$ and
849: for $N=20$ we get $C_{\mathrm{A}\mathrm{C}}=0.19$. It is
850: interesting to compare this results with the known finite-size
851: analysis scaling for the ground state energy of the Heisenberg
852: model \cite{manousakis}. The same scaling gives
853: $C_{\mathrm{A}\mathrm{C}}\approx {1 \over 6}+2N^{-3/2}$. Our
854: finite-size scaling, Fig.~\ref{fig8}(b), is in perfect agreement
855: with this result for $N\to \infty$ at $T=0$ and $B=0$.
856:
857: In the opposite limit of high magnetic fields, the vanishing
858: concurrence $C_{\rm AC}=0$, is observed for $B$ above the critical
859: value $B_c=4J'$ for all system sizes shown in Fig.~\ref{fig8}(a).
860: This result can be deduced also analytically. Since in a fully
861: polarized state $C_{\rm AC}=0$, this $B_c$ actually denotes a
862: transition from $S_1=N/2-1$ to $S_0=N/2$ ferromagnetic ground
863: state with energy $E_0=N(J'-B)/2$. The energy of the one-magnon
864: excitation above the ferromagnetic ground state is given by the
865: spin wave theory, which is in this case exact, as
866: $E_1=E_0-J'(2-\cos k_x a-\cos k_y a)+B$, where $(k_x,k_y)$ is the
867: magnon wave vector and $a$ denotes the lattice spacing. Evidently,
868: a transition to a fully polarized state occurs precisely at
869: $B_c=4J'$ at $(\pi/a,\pi/a)$ point in the one-magnon Brillouin
870: zone.
871: %
872:
873: \section{Summary}
874: \label{secV}
875:
876: The aim of this paper was to analyze and understand how concurrence
877: (and related entanglement) of qubit pairs (dimers) is affected by
878: their mutual magnetic interactions. In particular, we were interested
879: in a planar array of qubit dimers described by the Shastry-Sutherland
880: model. This model is suitable due to very robust ground state composed
881: of entangled qubit pairs which breaks down by increasing the
882: interdimer coupling. It is interesting to study both, the entanglement
883: between nearest and between next-nearest spins (qubits) at finite
884: temperature and magnetic field. The results are based on numerical
885: calculations using low-temperature Lanczos methods on lattices of 4, 8,
886: 16 and 20 sites with periodic boundary conditions.
887:
888: A comprehensive analysis of concurrence for various parameters
889: revealed two general conclusions:
890:
891: (1) For a weak coupling between qubit dimers, $J'<J'_c$, qubit pairs
892: are locally entangled in accordance with the local nature of the dimer
893: phase. This is due to a finite singlet-triplet gap (spin gap) in the
894: excitation spectrum that is a consequence of strong geometrical
895: frustration in magnetic couplings. The regime of fully entangled
896: neighbors perfectly coincides with the regime of finite spin gap as
897: presented in Fig.~\ref{fig9}. Calculated lines for various system
898: sizes $N$ in Fig.~\ref{fig9}(a) denote regions (shaded for $N=20$) in
899: the $(J'/J,B/J)$ plane where $C_{\mathrm{A}\mathrm{B}}=1$ at
900: $T=0$. In the lower panel [Fig.~\ref{fig9}(b)] the lines
901: represent the energy gap $E_1-E_{\mathrm{GS}}$ between the first
902: excited state with energy $E_1$ with total spin projection $S^z=1$ and the ground
903: state with energy $E_{\mathrm{GS}}$ and total
904: spin projection $S^z=0$, calculated for $B=0$. For $J'<J'_c$
905: (full lines) $E_1-E_{\mathrm{GS}}$ corresponds to the value of the
906: spin gap. With an increasing magnetic field the spin gap closes
907: (shaded region for $N=20$) and eventually vanishes at the
908: $C_{\mathrm{A}\mathrm{B}}=1$ border line. Shaded regions in
909: Figs.~\ref{fig9}(a),(b) therefore coincide. Note also that the results
910: for $N=16$ and 20 sites differ mainly in $J'_c$.
911:
912: As a consequence of finite spin gap and local character of
913: correlations it is an interesting observation that even $N=4$ results
914: as a function of temperature and magnetic field qualitatively
915: correctly reproduce $N=20$ results in the regime of $J'<J'_c$. The
916: main quantitative difference is in a renormalized value of $J'_c=J/2$
917: for $N=4$, as is evident from the comparison of
918: Figs.\ref{fig2},\ref{fig3} and Figs.\ref{fig5},\ref{fig6}. This
919: similarity of the results appears very useful due to the fact that
920: concurrence for tetrahedron-like systems ($N=4$) is given analytically
921: (Sec. \ref{secII}).
922:
923: (2) In the opposite, strong interdimer coupling regime, $J'>J'_c$, the
924: excitation spectrum is gapless and the concurrence between
925: next-nearest qubits, $C_{\mathrm{A}\mathrm{C}}$, exhibits a similar
926: behavior as in the antiferromagnetic Heisenberg model $J/J'\to 0$. Our
927: $B=0$ results coincide with the known result extrapolated to the
928: thermodynamic limit $C_{\mathrm{A}\mathrm{C}}\approx {1 \over 6}$. In
929: finite magnetic field and $T=0$ the concurrence vanishes at $B_c=4J'$
930: when the system becomes fully polarized (ground state with the total
931: spin $S=N/2$). However, at elevated temperatures the concurrence
932: increases due to excited singlet states and eventually drops to zero
933: at temperatures above $T_c\approx J'$.
934:
935: We can conclude with the observation that our analysis of concurrence
936: and related entanglement between qubit pairs was also found to be a
937: very useful measure for classifying various phases of the
938: Shastry-Sutherland model. As our numerical method is based on
939: relatively small clusters, we were unable to unambiguously determine
940: possible intermediate phases of the model in the regime $J' \sim
941: J'_c$, but we believe that concurrence will prove to be a useful probe
942: for the classification of various phases also in this regime using
943: alternative approaches. However, we were able to sweep through all
944: other dominant regimes of the parameters including finite temperature
945: and magnetic field.
946:
947: \section{Acknowledgments}
948:
949: The authors acknowledge J. Mravlje for useful discussions and the
950: support from the Slovenian Research Agency under Contract No. P1-0044.
951:
952:
953:
954: %
955: \begin{figure}
956: \begin{center}
957: \includegraphics[width=75mm,keepaspectratio]{Fig9.eps}
958: \end{center}
959:
960: \caption{(Color online) (a) Zero-temperature $C_{\mathrm{A}\mathrm{B}}=1$
961: region in the plane $(J'/J,B/J)$ for various $N$. (b) The
962: corresponding spin gap at $B=0$ (the energy of the lowest total
963: $S^z=1$ state relative to the ground state energy).}
964:
965: \label{fig9}
966: \end{figure}
967:
968:
969: \appendix
970: \section{Eigenenergies and eigenvectors for periodically coupled
971: two qubit dimers}
972: %
973: \noindent
974: Consider two qubit dimers coupled into a tetramer and described with
975: the Hamiltonian Eq.~(\ref{habcd}) and Fig.~\ref{fig1}(a). The model is
976: exactly solvable in the separate $\{S,S^z\}$ subspaces corresponding
977: to different values of the total spin $S$ and its $z$ component
978: $S^z$. Following the abbreviations for singlet and triplet states on
979: nearest-neighbor (dimer) sites $i$ and $j$,
980: %
981: \begin{eqnarray}
982: |s_{ij}\rangle &=& \frac{1}{\sqrt{2}}\, |\!\uparrow_i\downarrow_j -
983: \downarrow_i\uparrow_j \rangle,\nonumber\\
984: |t_{ij}^0\rangle &=& \frac{1}{\sqrt{2}}\, |\!\uparrow_i\downarrow_j +
985: \downarrow_i\uparrow_j \rangle,\nonumber\\
986: |t_{ij}^+\rangle &=& |\!\uparrow_i\uparrow_j \rangle,\nonumber\\
987: |t_{ij}^-\rangle &=& |\!\downarrow_i\downarrow_j \rangle,
988: \end{eqnarray}
989: %
990: the resulting eigenstates $|\phi_k\rangle$ and eigenenergies $E_k$
991: corresponding to the hamiltonian Eq.~(\ref{habcd}) are:
992: %
993: \allowdisplaybreaks
994: \begin{align}
995: S=0,&\ S^z=0:\nonumber\\
996: %
997: |\phi_1\rangle &= |s_{\rm AB}\rangle |s_{\rm CD}\rangle,\nonumber\\
998: E_1 &= -3J/2,\\[0.15cm]
999: |\phi_2\rangle &= \frac{1}{\sqrt{3}}\big(
1000: -|t_{\rm AB}^0\rangle |t_{\rm CD}^0\rangle +
1001: |t_{\rm AB}^+\rangle |t_{\rm CD}^-\rangle +
1002: |t_{\rm AB}^-\rangle |t_{\rm CD}^+\rangle\big)\nonumber\\
1003: E_2 &= J/2-4J'.\\\nonumber\\
1004: %
1005: %
1006: S=1,&\ S^z=-1:\nonumber\\
1007: %
1008: |\phi_3\rangle &= |s_{\rm AB}\rangle |t_{\rm CD}^-\rangle,\nonumber\\
1009: |\phi_4\rangle &= |t_{\rm AB}^-\rangle |s_{\rm CD}\rangle,\nonumber\\
1010: E_{3,4} &= -J/2-B,\\[0.15cm]
1011: |\phi_5\rangle &= \frac{1}{\sqrt{2}}\big(
1012: |t_{\rm AB}^0\rangle |t_{\rm CD}^-\rangle -
1013: |t_{\rm AB}^-\rangle |t_{\rm CD}^0\rangle\big),\nonumber\\
1014: E_5 &= J/2-2J'-B.\\\nonumber\\
1015: %
1016: %
1017: S=1,&\ S^z=0:\nonumber\\
1018: %
1019: |\phi_6\rangle &=|s_{\rm AB}\rangle |t_{\rm CD}^0\rangle,\nonumber\\
1020: |\phi_7\rangle &=|t_{\rm AB}^0\rangle |s_{\rm CD}\rangle,\nonumber\\
1021: E_{6,7} &= -J/2,\\[0.15cm]
1022: |\phi_8\rangle &= \frac{1}{\sqrt{2}}\big(
1023: |t_{\rm AB}^+\rangle |t_{\rm CD}^-\rangle -
1024: |t_{\rm AB}^-\rangle |t_{\rm CD}^+\rangle\big),\nonumber\\
1025: E_8 &= J/2-2J'.\\\nonumber\\
1026: %
1027: %
1028: S=1,&\ S^z=1:\nonumber\\
1029: %
1030: |\phi_9\rangle &= |s_{\rm AB}\rangle |t_{\rm CD}^+\rangle,\nonumber\\
1031: |\phi_{10}\rangle &= |t_{\rm AB}^+\rangle |s_{\rm CD}\rangle,\nonumber\\
1032: E_{9,10} &= -J/2+B,\\[0.15cm]
1033: |\phi_{11}\rangle &= \frac{1}{\sqrt{2}}\big(
1034: -|t_{\rm AB}^0\rangle |t_{\rm CD}^+\rangle +
1035: |t_{\rm AB}^+\rangle |t_{\rm CD}^0\rangle\big),\nonumber\\
1036: E_{11} &= J/2-2J'+B.\\\nonumber\\
1037: %
1038: %
1039: S=2,&\ S^z=-2:\nonumber\\
1040: %
1041: |\phi_{12}\rangle &= |t_{\rm AB}^-\rangle |t_{\rm CD}^-\rangle,\nonumber\\
1042: E_{12} &= J/2+2J'-2B.\\\nonumber\\
1043: %
1044: %
1045: S=2,&\ S^z=-1:\nonumber\\
1046: %
1047: |\phi_{13}\rangle &= \frac{1}{\sqrt{2}}\big(
1048: |t_{\rm AB}^0\rangle |t_{\rm CD}^-\rangle +
1049: |t_{\rm AB}^-\rangle |t_{\rm CD}^0\rangle\big),\nonumber\\
1050: E_{13} &= J/2+2J'-B.\\\nonumber\\
1051: %
1052: %
1053: S=2,&\ S^z=0:\nonumber\\
1054: %
1055: |\phi_{14}\rangle &= \frac{1}{2}\big(
1056: 2 |t_{\rm AB}^0\rangle |t_{\rm CD}^0\rangle +
1057: |t_{\rm AB}^+\rangle |t_{\rm CD}^-\rangle +
1058: |t_{\rm AB}^-\rangle |t_{\rm CD}^+\rangle\big),\nonumber\\
1059: E_{14} &= J/2+2J'.\\\nonumber\\
1060: %
1061: %
1062: S=2,&\ S^z=1:\nonumber\\
1063: %
1064: |\phi_{15}\rangle &= \frac{1}{\sqrt{2}}\big(
1065: |t_{\rm AB}^0\rangle |t_{\rm CD}^+\rangle +
1066: |t_{\rm AB}^+\rangle |t_{\rm CD}^0\rangle\big),\nonumber\\
1067: E_{15} &= J/2+2J'+B.\\\nonumber\\
1068: %
1069: %
1070: S=2,&\ S^z=2:\nonumber\\
1071: |\phi_{16}\rangle &= |t_{\rm AB}^+\rangle |t_{\rm CD}^+\rangle,\nonumber\\
1072: E_{16} &= J/2+2J'+2B.
1073: \end{align}
1074:
1075: \begin{thebibliography}{10}
1076: %
1077: \bibitem{nielsen01}M. A. Nielsen and I. A. Chuang, \emph{Quantum
1078: Information and Quantum Computation} (Cambridge University Press,
1079: Cambridge, 2001).
1080: %
1081: %\bibitem{shor94} P.W. Shor, Proc. 35th Annu. Symp. Foundations of
1082: %Computer Science, 124, IEEE Computer Society Press, Los Alamitos, 1994.
1083: %
1084: \bibitem{osterloh02}A. Osterloh, L. Amico, G. Falci, and R. Fazio,
1085: Nature \textbf{416}, 608 (2002).
1086: %
1087: \bibitem{bennett96}C.~H. Bennett, H.~J. Bernstein, S. Popescu, and B. Schumacher,
1088: Phys. Rev. A \textbf{53}, 2046 (1996); C.~H. Bennett, D.~P. DiVincenzo,
1089: J.~A. Smolin, and W.K. Wootters, \emph{ibid.} \textbf{54}, 3824 (1996).
1090: %
1091: \bibitem{hill97} S. Hill and W.~K. Wootters,
1092: Phys. Rev. Lett. \textbf{78}, 5022 (1997).
1093: %
1094: \bibitem{vedral97} V. Vedral, M.~B. Plenio, M.~A. Rippin, and P.~L. Knight,
1095: Phys. Rev. Lett. \textbf{78}, 2275 (1997).
1096: %
1097: \bibitem{wootters98} W. K. Wootters, Phys. Rev. Lett. \textbf{80}, 2245 (1998).
1098: %
1099: \bibitem{schliemann01}J. Schliemann, D. Loss, and A.~H. MacDonald, Phys. Rev. B
1100: \textbf{63}, 085311 (2001); J. Schliemann, J.~I. Cirac, M. Ku{\' s},
1101: M. Lewenstein, and D. Loss, Phys. Rev. A \textbf{64}, 022303 (2001).
1102: %
1103: \bibitem{ghirardi04}G.C. Ghirardi and L. Marinatto, Phys. Rev. A
1104: \textbf{70}, 012109 (2004).
1105: %
1106: \bibitem{eckert02}K. Eckert, J. Schliemann, G. Brus, and M. Lewenstein, Ann.
1107: Phys. \textbf{299}, 88 (2002).
1108: %
1109: \bibitem{gittings02}J.~R. Gittings and A. J. Fisher, Phys. Rev. A
1110: \textbf{66} 032305 (2002).
1111: %
1112: \bibitem{zanardi02}P. Zanardi, Phys. Rev. A \textbf{65}, 042101 (2002).
1113: %
1114: \bibitem{vedral03}V. Vedral, Cent. Eur. J. Phys. \textbf{2}, 289 (2003);
1115: D. Cavalcanti, M.~F. Santos, M.~O. TerraCunha, C. Lunkes, V. Vedral,
1116: Phys. Rev. A \textbf{72}, 062307 (2005).
1117: %
1118: \bibitem{criteria} D.~P.~DiVincenzo, {\it Mesoscopic Electron Transport,
1119: NATO Advanced Studies Institute, Series E: Applied Science}, edited by L.
1120: Kouwenhoven, G. Sch{\" o}n, and L. Sohn (Kluwer Academic, Dordrecht, 1997);
1121: cond-mat/9612126.
1122: %
1123: \bibitem{divincenzo05} D.~P.~DiVincenzo, Science \textbf{309}, 2173 (2005).
1124: %
1125: \bibitem{coish06} W.~A.~Coish~and~D.~Loss, cond-mat/0603444.
1126: %
1127: \bibitem{elzerman03} J.~M.~Elzerman, R.~Hanson, J.~S.~Greidanus,
1128: L.~H.~Willems van Beveren, S.~DeFranceschi, L.~M.~K.~Vandersypen, S.~Tarucha,
1129: and L.~P.~Kouwenhoven, Phys. Rev. B~\textbf{67}, 161308 (2003).
1130: %
1131: %---------------------------------------------------------------
1132: \bibitem{chen04} J.~C. Chen, A.~M. Chang, and M.~R. Melloch,
1133: Phys. Rev. Lett. \textbf{92}, 176801 (2004).
1134: %
1135: \bibitem{hatano05} T. Hatano, M. Stopa, and S. Tarucha, Science
1136: \textbf{309}, 268 (2005).
1137: %
1138: %---------------------------------------------------------------
1139: %
1140: \bibitem{petta05} J.~R.~Petta, A.~C.~Johnson, J.~M.~Taylor, E.~A.~Laird,
1141: A.~Yacoby, M.~D.~Lukin, C.~M.~Marcus, M.~P.~Hanson, and~A.~C.~Gossard,
1142: Science \textbf{309}, 2180 (2005).
1143: %
1144: \bibitem{nielsen00}M. A. Nielsen, Ph.D. thesis, University of New Mexico, 1998;
1145: quant-ph/0011036.
1146: %
1147: \bibitem{arnesen01}M. C. Arnesen, S. Bose, and V. Vedral,
1148: Phys. Rev. Lett. {\bf 87}, 017901 (2001).
1149: %
1150: \bibitem{wang01}X. Wang, Phys. Lett. A {\bf 281}, 101 (2001).
1151: %
1152: \bibitem{fine05}B. V. Fine, F. Mintert, and A.
1153: Buchleitner Phys. Rev. B {\bf 71}, 153105 (2005).
1154: %
1155: \bibitem{rmzb06}A. Ram{\v s}ak, J. Mravlje, R. {\v Z}itko, and
1156: J. Bon{\v c}a, Phys. Rev. B {\bf 74}, 241305(R) (2006).
1157: %
1158: \bibitem{yu02}T. Yu and J. H. Eberly, Phys. Rev. B {\bf 66}, 193306 (2002).
1159: %
1160: %---------------------------------------------------------------
1161: \bibitem{shastry81} B.~S. Shastry and B. Sutherland, Physica B
1162: \textbf{108}, 1069 (1981).
1163: %---------------------------------------------------------------
1164: %
1165: \bibitem{syljuasen03}O. F. Sylju\aa{}sen, Phys. Rev. A \textbf{68},
1166: 060301(R) (2003)
1167: %
1168: \bibitem{amico04}L. Amico, A. Osterloh, F. Plastina, R. Fazio, and
1169: G.M. Palma, Phys. Rev. A \textbf{69}, 022304 (2004).
1170: %
1171: \bibitem{osborne02}T. J. Osborne and M. A. Nielsen, Phys. Rev. A {\bf
1172: 66}, 032110 (2002).
1173: %
1174: \bibitem{roscilde04}T. Roscilde, P. Verrucchi, A. Fubini, S. Haas, and
1175: V. Tognetti, Phys. Rev. Lett. \textbf{93}, 167203 (2004).
1176: %
1177: \bibitem{roscilde05}T. Roscilde, P. Verrucchi, A. Fubini, S. Haas, and
1178: V. Tognetti, Phys. Rev. Lett. \textbf{94}, 147208 (2005).
1179: %
1180: \bibitem{larsson05}D. Larsson and H. Johannesson, Phys. Rev. Let. {\bf
1181: 95}, 196406 (2005).
1182: %
1183: \bibitem{gu04}Shi-Jian Gu, Shu-Sa Deng, You-Quan Li, and Hai-Qing Lin
1184: Phys. Rev. Lett. {\bf 93}, 086402 (2004).
1185: %
1186: \bibitem{deng05}S. S. Deng, S. J. Gu, and H. Q. Lin Phys. Rev. B {\bf 74},
1187: 045103 (2006).
1188: %
1189: \bibitem{legeza06}{\" O}. Legeza and J. S{\' o}lyom, Phys. Rev. Lett. {\bf 96},
1190: 116401 (2006).
1191: %
1192: \bibitem{verstraete04}F. Verstraete, M. Popp, and J. I. Cirac, Phys. Rev. Lett.
1193: {\bf 92}, 027901 (2004).
1194: %
1195: \bibitem{ramsak06} A. Ram\v{s}ak, I. Sega, and J. H. Jefferson
1196: Phys. Rev. A \textbf{74}, 010304(R) (2006).
1197: %
1198: \bibitem{werner89} R.~F. Werner, Phys. Rev. A \textbf{40}, 4277 (1989).
1199: %
1200: \bibitem{bose05}I. Bose and A. Tribedi, Phys. Rev. A {\bf 72}, 022314
1201: (2005).
1202: %
1203: %---------------------------------------------------------------
1204: % new - start
1205: %---------------------------------------------------------------
1206: \bibitem{smith91} R.~W. Smith and D.~A. Keszler, J. Solid State
1207: Chem. \textbf{93}, 430 (1991).
1208: %
1209: \bibitem{kageyama99}
1210: H. Kageyama, K. Yoshimura, R. Stern, N.~V. Mushnikov, K. Onizuka,
1211: M. Kato, K. Kosuge, C.~P. Slichter, T. Goto, and Y. Ueda,
1212: Phys. Rev. Lett. {\bf 82}, 3168 (1999).
1213: %
1214: \bibitem{miyahara03}
1215: S. Miyahara and K. Ueda, J. Phys.: Condens. Matter {\bf 15}, R327
1216: (2003); and references therein.
1217: %
1218: \bibitem{jorge03}
1219: { G.~A. Jorge, R. Stern, M. Jaime, N. Harrison, J. Bon\v ca, S. El
1220: Shawish, C.~D. Batista, H.~A. Dabkowska, and B.~D. Gaulin, Phys
1221: Rev. B \textbf{71}, 092403 (2005).}
1222: %
1223: \bibitem{elshawish1}
1224: S. El Shawish, J. Bon\v ca, C.~D. Batista, and I. Sega, Phys. Rev. B
1225: {\bf 71}, 014413 (2005).
1226: %
1227: \bibitem{elshawish2} S. El Shawish, J. Bon\v ca, and I. Sega,
1228: Phys. Rev. B {\bf 72}, 184409 (2005).
1229: %
1230: \bibitem{miyahara99}
1231: S. Miyahara and K. Ueda, Phys. Rev. Lett. {\bf 82}, 3701 (1999).
1232: %
1233: \bibitem{mueller00}
1234: E. M\"{u}ller-Hartmann, R. R. P. Singh, C. Knetter, and G.~S. Uhrig,
1235: Phys. Rev. Lett. {\bf 84}, 1808 (2000).
1236: %
1237: \bibitem{isacsson06}A. Isacsson and O. F. Sylju\aa{}sen, Phys. Rev. E
1238: {\bf 74}, 026701 (2006); and references therein.
1239: %
1240: \bibitem{onizuka00}
1241: K. Onizuka, H. Kageyama, Y. Narumi, K. Kindo, Y. Ueda, and T. Goto,
1242: J. Phys. Soc. Jpn. {\bf 69}, 1016 (2000).
1243: %
1244: \bibitem{momoi00}
1245: T. Momoi and K. Totsuka, Phys. Rev. B {\bf 61}, 3231 (2000).
1246: %
1247: \bibitem{misguich01}
1248: G. Misguich, Th. Jolicoeur, and S. M. Girvin, Phys. Rev. Lett., {\bf
1249: 87}, 097203 (2001).
1250: %
1251: \bibitem{ltlm}
1252: M. Aichhorn, M. Daghofer, H.~G. Evertz, and W. von der Linden,
1253: Phys. Rev. B {\bf 67}, 161103(R) (2003).
1254: %
1255: \bibitem{jj1} J. Jakli\v c and P. Prelov\v sek, Adv. Phys. \textbf{49}, 1
1256: (2000); Phys.\ Rev.\ Lett. \textbf{77}, 892 (1996); Phys. Rev. B
1257: \textbf{49}, 5065 (1994).
1258: %
1259: \bibitem {bonca02} J. Bon\v ca and P. Prelov\v sek, Phys. Rev. B 67,
1260: 085103 (2003).
1261: %
1262: \bibitem{wang012}X. Wang, Phys. Rev. A {\bf 64}, 012313 (2001); {\it
1263: ibid.} {\bf 66}, 044305 (2002).
1264: %
1265: \bibitem{cao05}M. Cao and S. Zhu, Phys. Rev. A {\bf 71}, 034311 (2005).
1266: %
1267: \bibitem{zhang05}G. F. Zhang and S. S. Li, Phys. Rev. A {\bf 72}, 034302 (2005).
1268: %
1269: \bibitem{manousakis} E. Manousakis, Rev. Mod. Phys. {\bf 63}, 1 (1991).
1270: %
1271: \end{thebibliography}
1272:
1273: \end{document}