1: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
2: %%
3: %% lastest version
4: %%
5: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
6: %\documentstyle[twocolumn,seceq]{jpsj}
7: \documentstyle[epsf,float,seceq,twocolumn]{jpsj}
8: %\documentstyle[short,epsf,float,seceq,twocolumn]{jpsj}
9: %\documentstyle[seceq,float,epsf,preprint]{jpsj}
10: %\documentstyle[seceq,float,epsf,preprint]{jpsj}
11:
12:
13: \font\eightmib=cmmib10 at 8pt\relax
14: \renewcommand{\mibs}[1]{\mbox{\eightmib #1}}
15:
16:
17: \author{Yoshihiro Takushima, Akihisa Koga, and Norio Kawakami}
18: \title{
19: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
20: Competing Spin-Gap Phases in a Frustrated Quantum Spin System
21: in Two Dimensions
22: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
23: }
24: \inst{Department of Applied Physics, Osaka University,
25: Suita, Osaka 565-0871}
26: \recdate{\today}
27: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28: \abst{
29: We investigate quantum phase transitions among the spin-gap phases
30: and the magnetically ordered phases in a two-dimensional
31: frustrated antiferromagnetic spin system,
32: which interpolates several important models
33: such as the orthogonal-dimer model as well as the model on the
34: 1/5-depleted square lattice.
35: By computing the ground state energy, the staggered susceptibility
36: and the spin gap by means of the series expansion method,
37: we determine the ground-state phase diagram and discuss
38: the role of geometrical frustration.
39: In particular, it is found that a RVB-type spin-gap phase
40: proposed recently for the orthogonal-dimer system is adiabatically
41: connected to the plaquette phase known for
42: the 1/5-depleted square-lattice model.
43: }
44: \kword{frustration, quantum phase transition, series expansion,
45: two dimensional spin system}
46: %\def\runtitle{Effects of inter-layer coupling for the orthogonal-dimer system}
47: %\def\runauthor{Akihisa {\sc Koga}}
48: \begin{document}
49: \sloppy
50: \maketitle
51: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
52: \section{Introduction}%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
53: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
54: Geometrical frustration plays an important role for the
55: spin-gap formation in a certain class of antiferromagnetic
56: quantum spin systems.
57: Two typical spin-gap compounds found in two dimension (2D) are
58: the transition metal oxides, $\rm SrCu_2(BO_3)_2$\cite{Kage} and
59: $\rm{CaV_{4}O_{9}}$.\cite{depE1}
60: In the former compound, the orthogonal-dimer structure of
61: the $\rm{Cu^{2+}}$ spins stabilizes the dimer phase
62: with the spin gap, which is properly described by the 2D Heisenberg model
63: on the square lattice with diagonal interactions
64: (Shastry-Sutherland model),\cite{S-S}
65: as claimed by Miyahara and Ueda.\cite{Ueda1}
66: For the model, it was pointed out that there should exist
67: an intermediate phase between the dimer
68: and the antiferromagnetic phases,\cite{Mila,KogaL,2trip}
69: implying that strong frustration due to the competing interactions
70: is quite important for understanding this model.
71: On the other hand, the latter spin-gap compound $\rm{CaV_{4}O_{9}}$
72: may be described by the 2D Heisenberg model
73: on the 1/5-depleted square lattice\cite{depT1}
74: or the meta-plaquette model.\cite{Metap1,Metap2}
75: The spin-gap phase in this compound
76: results from the plaquette-singlet formation,
77: which is further stabilized by the frustration effect
78: due to the next-nearest-neighbor interactions.
79:
80: An interesting feature common to the above systems is that
81: geometrical frustration due to the
82: competing interactions plays an essential role to
83: stabilize the disordered ground state. It is thus important
84: to deal with these systems in a unified framework to
85: clarify the role of frustration systematically.
86: In particular, since a RVB-type
87: spin-gap phase\cite{KogaL} proposed for the Shastry-Sutherland
88: model may be caused by geometrical frustration,
89: it is desired to further clarify the nature of this spin-gap phase.
90: This is also important from the experimental point of view, because
91: the compound $\rm SrCu_2(BO_3)_2$ is expected to be located around
92: the phase boundary between the dimer and RVB-type phases.
93:
94: Motivated by the above hot topics, we study the competition among
95: the spin-gap phases as well as the magnetically ordered phases
96: in a 2D frustrated quantum spin system on the square lattice
97: with some diagonal couplings. The basic idea in this paper is the
98: adiabatic continuation, which allows us to relate
99: apparently different states which belong to the same phase
100: at the fixed point. By using this model, we
101: systematically describe quantum phase transitions
102: for the orthogonal-dimer model
103: and the Heisenberg model on the 1/5-depleted square lattice,
104: where the ground-state phase diagram was already
105: discussed in detail.\cite{Ueda1,Mila,KogaL,2trip,KogaB,depT3}
106: In particular,
107: we show that a RVB-type spin-gap phase naturally emerges
108: in the Shastry-Sutherland model,
109: by observing the adiabatic evolution of the spin-gap state when
110: the competing interactions are varied.
111:
112: This paper is organized as follows.
113: In {\S 2}, we introduce a 2D frustrated spin model,
114: which allows us to treat the above spin-gap compounds systematically.
115: After briefly summarizing the series expansion method in {\S 3},
116: we investigate the first- and second-order quantum phase transitions
117: among the spin-gap phases and the magnetically ordered phases
118: in {\S 4}. We there determine the phase diagram and clarify
119: the role of geometrical frustration in this class of
120: quantum spin systems.
121: A brief summary is given in {\S 5}.
122:
123: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
124: \section{Model Hamiltonian}%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
125: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
126: We consider the Heisenberg model on the square lattice
127: with some diagonal couplings,
128: which is described by the Hamiltonian
129: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
130: \begin{eqnarray}
131: H&=&J\sum_{diagonal}{\bf S}_{i}
132: \cdot{\bf S}_{j}+
133: J'\sum_{square}{\bf S}_{i}
134: \cdot{\bf S}_{j},
135: \label{eq:model}
136: \end{eqnarray}
137: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
138: where ${\bf S}_{i}$ is the s = 1/2 spin operator at the {\it i}-th
139: site, and $J$, $J'_1$ and $J_2'$ represent the antiferromagnetic
140: exchange couplings. By putting $J$ on the orthogonal dimer
141: bonds (dashed lines), we then distribute the other
142: couplings $J'_{1}$ and $J'_{2}$
143: in two ways as shown in Figs. \ref{fig:md23}(a) and \ref{fig:md23}(b).
144: In the case (a), there is a diagonal bond $J$ in each plaquette
145: formed by the coupling $J'_1$ (bold line),
146: while in the case (b) there is no diagonal bond in shaded plaquettes.
147: What is interesting in this generalization is that we can
148: discuss several different models in the same framework.
149: Namely, both models of (a) and (b) are related to each other
150: via the Shastry-Sutherland model with $J'_{1}=J'_{2}$, and moreover they
151: are reduced to the well-known models in the limiting case of $J'_2=0$:
152: the orthogonal-dimer chain for (a)
153: and the spin system with a windmill structure for (b). Note that
154: the latter is topologically equivalent to the Heisenberg model
155: on the 1/5-depleted square lattice for the compound $\rm CaV_4O_9$.
156: Therefore, by using the above models we can systematically study
157: how the geometrical frustration in the 2D model affects
158: the generation of the spin gap.
159: For later convenience, we introduce the ratios $\alpha=J'_{1}/J$
160: and $k=J'_{2}/J'_{1}$.
161:
162:
163: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
164: \begin{figure}[htb]
165: \vspace{0.1cm}
166: \epsfxsize=7cm
167: \centerline{\epsfbox{model23cX.eps}}
168: \caption{Frustrated spin models on the square lattice with
169: some diagonal bonds.
170: The bold, thin and dashed lines represent the
171: coupling constants $J'_{1}, J'_{2}$
172: % = \lambda k J'_{1}$
173: and $J$.
174: The initial configuration of the isolated dimers (plaquettes)
175: for the series expansion is
176: indicated by the dashed lines (shaded squares).}
177: \label{fig:md23}%-----------------------------------------
178: \end{figure}
179: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
180:
181:
182: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
183: \section{Series Expansion Method}
184: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
185:
186: In order to study quantum phase transitions,
187: we make use of the series expansion method.\cite{Gel1}
188: This method is powerful especially for
189: quantum spin systems with frustration,
190: where quantum Monte Carlo simulations may suffer from sign problems.
191: In the present context, we wish to mention that this method was successfully
192: applied to the meta-plaquette system,\cite{Metap1,Metap2}
193: the orthogonal dimer system,\cite{KogaL,2trip,KogaB,Wei} etc.
194: To determine the phase diagram,
195: we consider two initial configurations in
196: the series expansion. The first one is a
197: plaquette-singlet configuration composed of the shaded
198: squares in Figs. \ref{fig:md23}(a) and \ref{fig:md23}(b).
199: In this case, we parameterize
200: the antiferromagnetic couplings labeled by
201: the bold, thin and dashed lines as
202: $J'_{1}, J'_{2}=\lambda kJ'_{1}$ and $J=\lambda J'_{1}/\alpha$,
203: where we have introduced an auxiliary series-expansion
204: parameter $\lambda$ in addition to the
205: dimensionless couplings $k$ and $\alpha$. The original model we
206: are interested in
207: is recovered by setting $\lambda=1$.
208: Note that a RVB-type spin-gap phase,\cite{KogaL}
209: which has the spatially-extended
210: disordered ground state, may be reached in this approach.
211: The second one is a dimer configuration, by which
212: we can describe the dimer phase. A convenient
213: parameterization in the dimer expansion is
214: $J(=1), J'_{1}=\lambda \alpha$ and $J'_{2}=\lambda k\alpha$.
215: In this way, these two different approaches enable
216: us to perform the series expansion
217: starting from isolated dimer and plaquette singlets ($\lambda=0$).
218:
219: To carry out the series expansion explicitly,
220: the original Hamiltonian eq.(\ref{eq:model}) is divided
221: into two parts,
222: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
223: \begin{eqnarray}
224: H&=&H_{0}+\lambda H_{1}\nonumber\\
225: &=&x\left[\sum{\bf S}_{i}\cdot{\bf S}_{j}+
226: \lambda\sum\Gamma_{ij}{\bf S}_{i}\cdot{\bf S}_{j}\right],\label{eq:model1}
227: \end{eqnarray}
228: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
229: where $x=J$ $(J'_{1})$ and $\Gamma_{ij}=\alpha$ or $k\alpha$ $(k$ or
230: $\alpha^{-1})$ for the dimer (plaquette) expansion.
231: The first term $H_{0}$ is the unperturbed Hamiltonian
232: which stabilizes the isolated dimer or plaquette singlets, whereas
233: the perturbed Hamiltonian $H_{1}$
234: labeled by $\lambda$ connects these isolated
235: dimer or plaquette singlets, yielding a 2D frustrated
236: spin system.
237: We compute the staggered susceptibility $\chi_{AF}$, the spin-triplet
238: excitation energy $E({\bf q})$, and the ground state energy $E_{g}$ as
239: a power series in $\lambda$.
240: To estimate the susceptibility, we introduce the Zeeman term
241: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
242: \begin{eqnarray}
243: H' &=& h \left[\sum_{i\in A}{\bf S}^{z}_{i}-
244: \sum_{i\in B}{\bf S}^{z}_{i}\right],
245: \label{eq:model2}
246: \end{eqnarray}
247: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
248: where $h$ is the staggered magnetic field and $A (B)$ denotes one of
249: the two sublattices.
250:
251: We determine the quantum phase transition point in three ways:
252: (i) comparison of the ground state energy, (ii)
253: divergence of the staggered susceptibility, (iii)
254: vanishing point of the spin gap.
255: The latter two schemes may be applied to the second-order
256: phase transition, while the first one is more effective
257: to study the first-order phase transition.
258: In order to obtain the accurate phase boundary,
259: we further use the Dlog Pad$\rm{\acute{e}}$
260: approximants\cite{Pade} for the calculated results up to the finite
261: order in $\lambda$.
262: We also make use of the {\it{biased}} Pad$\rm{\acute{e}}$
263: approximants,\cite{Pade}
264: in which the critical exponents for our 2D quantum spin
265: model are assumed to take those expected for the 3D classical
266: Heisenberg model.\cite{3Dcl}
267: The phase-transition point, $\lambda_{c}$, is then determined by the
268: formula $\chi_{AF} \sim (\lambda_{c}-\lambda)^{-\gamma}$ and
269: $\Delta \sim (\lambda_{c}-\lambda)^{\nu}$ with the critical exponents
270: $\gamma = 1.4$ and $\nu = 0.71.$\cite{Crit}
271:
272: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
273: \section{Quantum Phase Transitions}
274: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
275: We now discuss quantum phase transitions
276: when the competing exchange couplings are varied.
277: There are three parameters in the perturbation term,
278: $\lambda, \alpha$ and $k$.
279: We first study the model given by
280: Fig. \ref{fig:md23}(a), which continuously connects the
281: 2D Shastry-Sutherland model and the 1D orthogonal-dimer chain, and
282: then move to the model (b), which includes the
283: Heisenberg model on the 1/5 depleted square lattice.
284:
285: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
286: \subsection{Adiabatic evolution from the orthogonal-dimer chain}
287: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
288:
289: Let us start with the system shown in Fig. \ref{fig:md23}(a).
290: Note that this spin system is reduced to
291: the orthogonal-dimer chain for $k=0$,
292: which was already studied in detail:
293: when the ratio $\alpha$ of the competing exchange couplings is varied,
294: the first-order quantum phase transition
295: between the spin-gap phases with the dimer and the plaquette structures
296: occurs at $\alpha_{c1}=0.81900$.\cite{Ivanov,KogaB}
297: The merit to study this system is that we can check how the
298: known spin-gap states in the chain system
299: adiabatically evolve when the system approaches
300: the 2D Shastry-Sutherland model.
301:
302: We first discuss the effect of interchain coupling $k$
303: on the first-order phase transition between two
304: spin-gap phases
305: by computing the ground state energy $E_{g}$.
306: Note that the energy for the dimer phase is unchanged
307: even on the introduction of $k$ due to the
308: orthogonal-dimer structure.\cite{S-S} For the plaquette phase, by
309: choosing the isolated plaquettes shown by shaded squares
310: in Fig. \ref{fig:md23}(a) as an initial configuration,
311: we compute the ground state energy $E_{g}$ up to the sixth order
312: in $\lambda$ with $\alpha$ and $k$ being fixed.
313: By applying the first-order inhomogeneous differential method to
314: the obtained series,
315: we then estimate the ground state energy for the plaquette phase
316: by setting $\lambda=1$.
317: The energy estimated for $k=0.1$, 0.5 and 0.9 is
318: shown in Fig. \ref{fig:pladene}.
319: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
320: \begin{figure}[htb]
321: \vspace{0.1cm}
322: \epsfxsize=8cm
323: \centerline{\epsfbox{pladene0951.eps}}
324: \caption{Ground state energy per site as a function of $\alpha = J'_{1}/J$
325: (for $\lambda=1$).
326: The flat line ($E_{g}/JN$ = -3/8) is the energy
327: for the exact dimer state. The dotted, dashed and solid
328: lines denote the energy at $\lambda=1$ obtained
329: by the plaquette expansion for $k=0.1$, 0.5 and 0.9, respectively.}
330: \label{fig:pladene}%-----------------------------------------
331: \end{figure}
332: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
333: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
334: \begin{figure}[htb]
335: \vspace{0.1cm}
336: \epsfxsize=7cm
337: \centerline{\epsfbox{pladphase1.eps}}
338: \caption{Phase diagram for the 2D frustrated spin system
339: shown in Fig. \ref{fig:md23}(a).
340: The phases I, II and III represent the dimer,
341: plaquette and magnetically ordered phases, respectively.
342: The dotted line corresponds to the 2D Shastry-Sutherland model and
343: the dot-dashed lines to the asymptotic phase boundary for large $\alpha$.
344: The cross indicates the parameters deduced for
345: ${\rm SrCu_{2}(BO_{3})_{2}} $.\cite{Ueda2}}
346: \label{fig:pladphase}%-----------------------------------------
347: \end{figure}
348: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
349: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
350: It is seen that the introduction of $k$
351: prefers the plaquette phase, shifting
352: the first-order transition point
353: $\alpha_{c1}$ to the left in Fig. \ref{fig:pladene}.
354: From these analyses,
355: we determine the phase boundary between the dimer phase and
356: the plaquette phase in the 2D
357: parameter space, which is shown in Fig. \ref{fig:pladphase}.
358: We have checked that the present estimate of
359: the critical point for $k=0$ reproduces the
360: above-mentioned value of $\alpha_{c1}$
361: fairly well. It is seen that
362: the phase boundary is not so sensitive to $k$ for small $k$,
363: because the energy for the plaquette phase decreases
364: quadratically in $k$.
365:
366: It is to be noted that these two competing spin-gap phases are
367: continuously connected to those in the Shastry-Sutherland model $(k=1)$,
368: with the nature of the first-order transition being
369: unchanged.\cite{KogaL} To check the validity of the obtained results,
370: we have also performed the plaquette expansion
371: starting from another initial
372: configuration, where an isolated plaquette is formed by
373: four spins shown as 1, 2, 3 and 4 in Fig. \ref{fig:md23}(a).
374: From the ground state energy computed up to the seventh order, we
375: confirm that the phase boundaries obtained by two distinct ways
376: are indeed consistent within our accuracy.
377:
378: To complete our discussions on this model,
379: it is necessary to discuss whether the system is driven to
380: the magnetically ordered phase with the increase of
381: the interchain coupling. To this end,
382: we calculate the staggered susceptibility $\chi_{AF}$ and the spin
383: gap $\Delta$ by means of the plaquette expansion up to the fourth
384: and the fifth order in $\lambda$, respectively.
385: We show the results obtained for $k=0.9$ and 0.8
386: in Fig. \ref{fig:pladgs}. It is seen that two lines determined
387: from different quantities provide
388: consistent values even around $\lambda=1$, implying
389: that the phase boundary may be obtained rather accurately for the
390: original model ($\lambda=1$) although our calculation is
391: restricted to the lower-order series expansion. From the
392: above analyses, we end up with the phase
393: boundary shown in Fig. \ref{fig:pladphase},
394: which separates the plaquette phase II and
395: the magnetically ordered phase III.
396: Note that for large $\alpha$, the boundary between II and III
397: approaches the correct asymptotic line,
398: which was estimated as $k_c=0.56$.\cite{Neelb}
399: This also supports that the present calculation
400: gives the reliable results for the phase diagram.
401:
402: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
403: \begin{figure}[htb]
404: \vspace{0.1cm}
405: \epsfxsize=8cm
406: \centerline{\epsfbox{pladgs098.eps}}
407: \caption{Second-order phase transition between the plaquette
408: phase and the antiferromagnetically ordered phase
409: for the 2D spin system given by
410: Fig. \ref{fig:md23}(a) (for $k=0.9$ and 0.8).
411: The solid (dashed) line indicates the phase boundary obtained
412: by biased [3/2] ([2/2]) Pad$\rm{\acute{e}}$ approximants for
413: the spin gap (the staggered susceptibility).}
414: \label{fig:pladgs}%-----------------------------------------
415: \end{figure}
416: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
417:
418: In the obtained phase diagram
419: shown in Fig. \ref{fig:pladphase},
420: we remark again that two competing spin-gap states in
421: the orthogonal dimer chain\cite{Ivanov,KogaB} are
422: continuously connected to those in
423: Shastry-Sutherland model. In particular, a RVB-like
424: intermediate spin-gap phase \cite{KogaL} proposed for the
425: Shastry-Sutherland model naturally emerges via the
426: adiabatic continuation of
427: the plaquette phase in the orthogonal-dimer chain.
428:
429:
430: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
431: \subsection{Adiabatic evolution from the 1/5-depleted square lattice}
432: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
433:
434: We next deal with the spin system with a slightly different
435: structure shown in Fig. \ref{fig:md23} (b).
436: Recall that when $k=1$ ($k=0$), this model with $\lambda=1$
437: is reduced to the Shastry-Sutherland model (1/5-depleted square lattice
438: model). We here study how the spin-gap phase in the
439: orthogonal-dimer model for ${\rm SrCu_{2}(BO_{3})_{2}}$
440: is related to that in the 1/5-depleted square-lattice
441: model for $\rm{CaV_{4}O_{9}}$, when the
442: frustrating interaction $k$ is continuously varied.
443: We carry out the dimer (plaquette) expansion
444: for the ground-state energy, the staggered susceptibility
445: and the spin gap. By
446: performing similar asymptotic analyses mentioned above,
447: the phase diagram is determined, which
448: is shown in Fig. \ref{fig:depphase}. The detail of calculation
449: will be described below for each case.
450:
451: Let us start our discussions with the case of $k=0$, i.e.
452: the 1/5-depleted square lattice model, which
453: was already studied in connection
454: with $\rm{CaV_{4}O_{9}}$.\cite{depT3}
455: In this case, as the ratio of the exchange interactions
456: $\alpha$ is changed, the dimer phase I first undergoes
457: the second-order phase transition to the antiferromagnetically ordered
458: phase IV at the critical point $\alpha_{c2}^{l}=0.590(5)$,
459: and this ordered phase further shows the second-order phase
460: transition to the plaquette phase II at $\alpha_{c2}^{u}=1.09(1)$.
461: The critical points obtained by the
462: present method agree fairly well with those of
463: quantum Monte Carlo simulations.\cite{depT3}
464: Note that the antiferromagnetically ordered phase on
465: 1/5-depleted square lattice labeled by IV
466: is different from that on the square lattice indicated by
467: III.
468:
469: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
470: \begin{figure}[htb]
471: \vspace{0.1cm}
472: \epsfxsize=7cm
473: \centerline{\epsfbox{depphase1.eps}}
474: \caption{Phase diagram for the 2D frustrated spin system
475: shown in Fig. \ref{fig:md23}(b).
476: The phase I and II represent the dimer and the plaquette phase.
477: The phase III (IV) represents the antiferromagnetically ordered
478: phase on the square lattice (1/5-depleted square lattice).
479: The dotted line indicates the 2D Shastry-Sutherland model and
480: the dot-dashed lines the asymptotic phase boundary for large $\alpha$.
481: The crosses + and $\times$
482: indicate the location of
483: ${\rm SrCu_{2}(BO_{3})_{2}} $\cite{Ueda2} and
484: ${\rm CaV_{4}O_{9}}$,\cite{Metap1} respectively.}
485: \label{fig:depphase}%-----------------------------------------
486: \end{figure}
487: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
488:
489: As the coupling $k$ is increased, the spin-gap
490: phases become more stable while
491: the magnetic correlation for the phase IV is
492: suppressed, which is mainly due to the frustrating
493: nature of the interaction $k$. It is seen that the competition
494: among three phases persists up to
495: the multicritical point $(\alpha, k\alpha)\simeq (0.72, 0.32)$
496: beyond which the magnetically ordered phase IV
497: disappears. Thus for larger $k$
498: the dimer state directly undergoes the first-order
499: phase transition to the plaquette spin-gap phase.
500: Note that the dimer phase here does not have the exact eigenstate
501: except for the orthogonal-dimer case $k=1$,
502: which is contrasted to the model (a).
503: We thus have estimated the ground state energy
504: both of the dimer and the plaquette phases in this case.
505: For reference, we show in Fig. \ref{fig:depene09}
506: the ground-state energy calculated for $k=0.9$, where
507: the dimer (plaquette) expansion has been first performed up to the
508: eighth (seventh) order in $\lambda$,
509: and then the first-order inhomogeneous differential method
510: has been used to obtain the values at $\lambda=1$.
511: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
512: \begin{figure}[htb]
513: \vspace{0.1cm}
514: \epsfxsize=8cm
515: \centerline{\epsfbox{depeneI.eps}}
516: \caption{Ground state energy per site as a function of
517: $\alpha=J'_{1}/J$ (for $\lambda=1$ and $k=0.9$) for the
518: model shown in Fig. \ref{fig:md23}(b).
519: The solid (dashed) line indicates the energy
520: for the dimer (plaquette) phase.
521: We also show the results obtained by the exact diagonalization
522: for $N=24$ sites with a periodic boundary condition as open circles.}
523: \label{fig:depene09}%-----------------------------------------
524: \end{figure}
525: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
526: To confirm the validity of our series expansion, we have also
527: performed the exact diagonalization
528: for the system with $N=24$ sites with a periodic boundary condition,
529: which is found to be consistent with the series-expansion results.
530: The phase boundary thus obtained between two spin-gap phases
531: is drawn in Fig. \ref{fig:depphase}.
532: We note that our series expansion
533: turns out to be not so accurate around the tricritical point, so that
534: the phase boundary around there is shown as the dashed line.
535:
536: When the coupling $k$ is further increased, the
537: plaquette phase
538: shows the instability to an antiferromagnetically
539: ordered phase III on the square lattice. To study this instability
540: from II to III, we have performed the plaquette expansion for
541: the staggered susceptibility (spin gap)
542: up to the fourth (fifth) order, and determined the phase boundary
543: using the biased Pad\'e approximants, which is
544: shown in Fig. \ref{fig:depphase}.
545:
546: Consequently, we arrive at the phase diagram
547: shown in Fig. \ref{fig:depphase}, in which
548: the parameters deduced for
549: the spin-gap compounds $\rm CaV_4O_9$ and $\rm SrCu_2(BO_3)_2$
550: are also indicated. It is clearly seen how the 2D Heisenberg
551: model on the 1/5-depleted square lattice
552: is connected to the Shastry-Sutherland model.
553: It is particularly interesting to notice
554: that a RVB-type spin gap state proposed
555: for the Shastry-Sutherland model is adiabatically connected to a
556: plaquette singlet state for $\rm CaV_4O_9$.
557:
558: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
559: \subsection{Excitation spectrum}
560: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
561:
562: We have seen that the spin-gap phases in the Shastry-Sutherland
563: model are continuously connected to the spin-gap phases
564: known for two different models. According to this relationship,
565: it may be expected that the Shastry-Sutherland model shares some
566: of its characteristic features with the latter models. As an example,
567: we here show that the spin-triplet excitation spectrum for the
568: RVB-type phase in the Shastry-Sutherland model
569: indeed exhibits a similar behavior expected for the
570: plaquette phase in the orthogonal-dimer chain.
571: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
572: \begin{figure}[htb]
573: \vspace{0.1cm}
574: \epsfxsize=7cm
575: \centerline{\epsfbox{gap.eps}}
576: \caption{Spin-triplet excitation energy
577: $\Delta/J$ for (a) Shastry-Sutherland model and
578: (b) orthogonal-dimer chain as a function of $\alpha(=J'_{1}/J)$.
579: In (a), a triplet excitation spectrum \cite{Wei} over the dimer state
580: is plotted for $\alpha<\alpha_{c1}$, while two lowest triplet
581: excitations are shown for the RVB-type phase
582: $\alpha_{c1}<\alpha<\alpha_{c2}$. In (b), similar plot is
583: done for the dimer phase ($\alpha<\alpha_c$) and also for
584: the plaquette phase ($\alpha_c <\alpha$).\cite{KogaB}
585: The meaning of the solid, dashed and dotted lines
586: is mentioned in the text.}
587: \label{fig:gap}%-----------------------------------------
588: \end{figure}
589: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
590: In Fig. \ref{fig:gap}(a), we show the spin-triplet
591: excitation spectrum calculated for the 2D Shastry-Sutherland model.
592: In contrast to the dimer phase, where
593: the lowest triplet excitation is an almost-localized
594: triplet over the dimer ground state,\cite{Ueda1} we find that
595: the lowest-triplet excitation in the RVB-type phase
596: shows a more complicated feature; two triplet
597: excitations show the level-crossing around value of $\alpha' \simeq 0.76$.
598: It is not easy to clearly see the origin
599: of this level-crossing because our calculation for the 2D model is
600: restricted to lower-order perturbation. Fortunately,
601: we can exploit the idea of the adiabatic continuation
602: by continuously changing the system to the
603: orthogonal-dimer chain.
604:
605: In the orthogonal-dimer chain, a similar level-crossing was
606: recently observed,\cite{KogaB} as shown in Fig. \ref{fig:gap}(b).
607: In this case, the nature of two different excitations in the
608: plaquette phase is well understood. Namely, in the lowest
609: excitation denoted by the dotted
610: line in the region of $\alpha>\alpha'(\simeq 0.87)$
611: is a triplet excitation that simply breaks a plaquette singlet.
612: On the other hand in the region $\alpha_{c1}<\alpha<\alpha'$,
613: the lowest-energy excited state shown by the dashed line
614: is described by a four-fold degenerate level
615: without dispersion,\cite{KogaB}
616: in which two diagonal spins on a square
617: plaquette becomes completely free,
618: giving rise to the four-fold degeneracy.\cite{KogaB}
619: The emergence of this peculiar excitation reflects the fact that the
620: system is located closely to the first-order phase transition point.
621: In other words, this level crossing is due to the strong
622: frustration effect around the transition point.
623: When the interchain coupling $k$ is introduced, the above two-types
624: of excitations are continuously
625: changed to those in the Shastry-Sutherland model, as
626: shown in Fig. \ref{fig:gap}(a). In particular,
627: four-fold degenerate states in the chain system
628: split into singlet and triplet states, and the latter
629: gives the lowest-excited state shown by the dashed
630: line in the Shastry-Sutherland model in Fig. \ref{fig:gap}(a).
631: In this way, the characteristic level-crossing found in the
632: 1D system persists even for the 2D system, from which
633: we can say that the unusual level-crossing
634: found in the Shastry-Sutherland model reflects
635: the strong frustration around the
636: first-order phase transition point.
637:
638: %%Note that the the vanishing excitation energy at $\alpha_{c2}$
639: %%is due to the transition to the magnetically ordered state,
640: %%which does not occur in the 1D system as seen
641: %%from Fig.\ref{fig:gap}(b).
642:
643:
644:
645:
646:
647: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
648: \section{Summary}%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
649: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
650:
651: We have studied the ground-state phase diagram
652: for a frustrated quantum spin model on the square lattice, which
653: systematically describes the systems for the
654: spin-gap compounds such as $\rm SrCu_2(BO_3)_2$ and $\rm CaV_4O_9$.
655: By calculating the ground state energy, the staggered
656: susceptibility and the spin gap by means of the series expansion method,
657: we have determined the phase diagram, and clarified
658: the nature of the associated quantum phase transitions.
659: One of the main purpose of the present study is to uncover
660: the origin of the intermediate
661: spin-gap state proposed for the Shastry-Sutherland model, which
662: has a spatially extended nature.
663: Starting from two types of the well-studied models, we have
664: observed how this intermediate phase emerges in the phase diagram.
665: Namely, with the use of the simple orthogonal-dimer chain model,
666: it has been checked that the
667: first-order phase transition between the dimer and RVB-type phases
668: in the Shastry-Sutherland model has the same origin as in the transition
669: between the dimer and plaquette phases known for the chain system.
670: Further analysis based on the second model has allowed us to
671: show that the RVB-type intermediate state is adiabatically
672: connected to the plaquette state known for the 1/5-depleted square
673: lattice model, which describes the essential properties of
674: $\rm CaV_4O_9$. We have also shown that
675: the level-crossing of the triplet excitations in the Shastry-Sutherland model
676: is understood well in terms of their counterparts in the
677: 1D orthogonal-dimer chain model.
678:
679: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
680: %%%%%%% new sentence %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
681: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
682: These systematic investigations based on the adiabatic
683: continuation support that the intermediate
684: spin-gap phase for the Shastry-Sutherland model
685: may be stabilized by geometrical frustration
686: induced by the competing interactions. However,
687: we wish to mention that there still remain unresolved
688: problems for the phase diagram of the Shastry-Sutherland model.
689: For example,
690: Weihong {\it et al.} have shown by means of the Ising expansion\cite{Wei}
691: that the system has the finite staggered magnetization down to
692: $(J'/J)\simeq 0.71$,
693: which may contradict the fact that the spin gap phase
694: is extended up to $(J'/J)\simeq 0.86 $. Although we have checked
695: that the ground state energy of the
696: present spin-gap phase for $(J'/J) < 0.86 $
697: is indeed lower than that of Weihong
698: {\it et al.}, this apparent inconsistency
699: for the magnetization should be resolved in the future study.
700: Also, Knetter {\it et al.} have recently claimed that the instability of
701: the two-magnon excitation in the dimer phase may occur at $(J'/J)=0.630(5)$.
702: \cite{2trip} Although it may not be
703: clear whether this indeed triggers a first-order phase
704: transition to some new phase,\cite{2trip} it is an important
705: open problem to answer the above points for the phase diagram
706: of the Shastry-Sutherland
707: model. This is now under consideration.
708:
709: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
710: %\section*{acknowledgments}%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
711: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
712: The work is partly supported by a
713: Grant-in-Aid from the Ministry of Education, Science, Sports, and Culture.
714: A.K. is supported by the Japan Society for the Promotion of Science.
715: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
716: \begin{thebibliography}{99}%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
717: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
718:
719: %%%%%% BrCu2(BO3)2 %%%%%%%%%%%%%%%%%
720: \bibitem{Kage}
721: H. Kageyama, K. Yoshimura, R. Stern, N. V. Mushnikov,
722: K. Onizuka, M. Kato, K. Kosuge, C. P. Slichter, T. Goto and
723: Y. Ueda,
724: Phys. Rev. Lett. {\bf 82} (1999) 3168.
725:
726: %%%%%% 1/5-dep Ex %%%%%%%%%%%%%%%%%
727: \bibitem{depE1}
728: S. Taniguchi, Y. Nishikawa, Y. Yasui, Y. Kobayashi,
729: M. Sato, T. Nishioka, M. Kontani and K. Sano,
730: J. Phys. Soc. Jpn. {\bf 64} (1995) 2758.
731:
732: %%%%%% S-S model %%%%%%%%%%%%%%%%%
733: \bibitem{S-S}
734: B. S. Shastry and B. Sutherland,
735: Physica {\bf 108B} (1981) 1069.
736:
737: %%%%%% S-S ueda1 %%%%%%%%%%%%%%%%%
738: \bibitem{Ueda1}
739: S. Miyahara and K. Ueda,
740: Phys. Rev. Lett. {\bf 82} (1999) 3701.
741:
742: %%%%%% S-S ueda1 %%%%%%%%%%%%%%%%%
743: \bibitem{Mila}
744: M. Albrecht and F. Mila,
745: Europhys. Lett. {\bf 34} (1996) 145.
746:
747: %%%%%% S-S and pla-chain %%%%%%%%%%%%%%%%%
748: \bibitem{KogaL}
749: A. Koga, and N. Kawakami,
750: Phys. Rev. Lett. {\bf 84} (2000) 4461.
751: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
752:
753: %%%%%%%%%%%%%%%%%%%%%
754: \bibitem{2trip}
755: C. Knetter, A. B$\rm{\ddot{u}}$hler, E. M.-Hartmann
756: and G. S. Uhrig,
757: Phys. Rev. Lett. {\bf 85} (2000) 3958.
758: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
759:
760:
761: %%\bibitem{Wei}
762: %%Z. Weihong, C. J. Hamer and J. Oitmaa,
763: %%Phys. Rev. {\bf B60} 6608 (1999)
764: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
765: %%\bibitem{Mul}
766: %%E. M${\rm\ddot{u}}$ller Hartmann, R. R. P. Singh,
767: %%C. Knetter and G. S. Uhrig,
768: %%Phys. Rev. Lett. {\bf 84} 1808 (2000).
769:
770: %%%%%% 1/5-dep T Ueda %%%%%%%%%%%%%%%%%
771: \bibitem{depT1}
772: K. Ueda, H. Kontani, M. Sigrist and P. A. Lee,
773: Phys. Rev. Lett. {\bf 76} (1996) 1932;
774: %\bibitem{depT2}
775: N. Katoh and M. Imada,
776: J. Phys. Soc. Jpn. {\bf 64} (1995) 4105.
777:
778: %%%%%% critical exponent %%%%%%%%%%%%%%%%%
779: \bibitem{Metap1}
780: Y. Fukumoto and A. Oguchi,
781: J. Phys. Soc. Jpn. {\bf 67} (1998) 2205.
782: \bibitem{Metap2}
783: Z. Weihong, J. Oitmma and C. J. Hamer,
784: Phys. Rev. B {\bf 58} (1998) 14147.
785:
786: %%%%%% Multiparticle excitation %%%%%%%%%%%%%%%%%
787: %\bibitem{2trip}
788: %C. Knetter, A. B$\rm{\ddot{u}}$hler, E. M.-Hartmann
789: %and G. S. Uhrig,
790: %cond-mat / 0005322.
791: %;S. Trebst, H. Monien, C. J. Hamer,
792: %Z. Weihong and R. R. P. Singh,
793: %cond-mat / 0007192
794: %%%%%% pla-chain Koga %%%%%%%%%%%%%%%%%
795: %%%%%%%%%%%%%%%%%%%%%%%%%%%
796: \bibitem{KogaB}
797: A. Koga, K. Okunishi and N. Kawakami,
798: Phys. Rev. B {\bf 62} (2000) 5558.
799:
800: %%%%%% 1/5-dep T Ueda %%%%%%%%%%%%%%%%%
801: \bibitem{depT3}
802: M. Troyer, H. Kontani and K. Ueda,
803: Phys. Rev. Lett. {\bf 76} (1996) 3822.
804: %%%%%% Gel 1 %%%%%%%%%%%%%%%%%
805: \bibitem{Gel1}
806: R. R. P. Singh, M. P. Gelfand and D. A. Huse,
807: Phys. Rev. Lett. {\bf 61} (1988) 2484.
808: %%%%%%%%%%%%%%%%%%%%%%%
809: \bibitem{Wei}
810: Z. Weihong, C. J. Hamer and J. Oitmma,
811: Phys. Rev. B {\bf 60} (1999) 6608.
812: %%%%%% Pade-ap %%%%%%%%%%%%%%%%%
813: \bibitem{Pade}
814: A. J. Guttmann, in {\it{Phase Transitions and Critical Phenomena}},
815: edited by C. Domb and J. L. Lebowitz (Academic, New York, 1989)
816: Vol. 13.
817:
818: %%%%%% 3D classical %%%%%%%%%%%%%%%%%
819: \bibitem{3Dcl}
820: S. Chakravarty, B. I. Halperin and D. R. Nelson,
821: Phys. Rev. Lett. {\bf 60} (1988) 1057;
822: Phys. Rev. B {\bf 39} (1989) 2344.
823:
824: %%%%%% critical exponent %%%%%%%%%%%%%%%%%
825: \bibitem{Crit}
826: M. Ferer and A. Hamid-Aidinejad,
827: Phys. Rev. B {\bf 34} (1986) 6481.
828:
829:
830: %%%%%% %%%%%%%%%%%%%%%%%
831: \bibitem{Ivanov}
832: N. B. Ivanov and J. Richter,
833: Phys. Lett. {\bf 232A} (1997) 308;
834: J. Richter, N. B. Ivanov and J. Schulenburg,
835: J. Phys. Condence Matt. {\bf 10} (1998) 3635.
836:
837: %%%%%% SrCu ueda1 %%%%%%%%%%%%%%%%%
838: \bibitem{Ueda2}
839: S. Miyahara and K. Ueda,
840: J. Phys. Soc. Jpn. Suppl. B {\bf 69} (2000) 72.
841: %cond-matt / 0004260.
842:
843: %%%%%% Square lattice 1 %%%%%%%%%%%%%%%%%
844: %\bibitem{Sq1}
845: %S. Chakravarty, B. I. Halpelin and D. R. Nelson,
846: %Phys. Rev. Lett. {\bf 60} 1057 (1988);
847: %Phys. Rev. {\bf B39} 2344 (1989);
848: %N. Read and S. Sachdev,
849: %Phys. Rev. Lett. {\bf 62} 1694 (1989);
850: %{\bf 66} 1773 (1991).
851: %%%%%% Square lattice 2 %%%%%%%%%%%%%%%%%
852: %\bibitem{Sq2}
853: %J. D. Reger and A. P. Young,
854: %Phys. Rev. {\bf B37} 5978 (1988).
855:
856: %%%%%% kondo lat %%%%%%%%%%%%%%%%%
857: %\bibitem{Kond}
858: %Z.-P. Shi, R. R. P. Singh, M. P. Gelfand and Z. Wang,
859: %Phys. Rev. B {\bf 51}, R15630 (1995).
860:
861: %%%%%% Bila %%%%%%%%%%%%%%%%%
862: %\bibitem{Bila}
863: %K. Hida,
864: %J. Phys. Soc. Jpn. {\bf 61}, 1013 (1992);
865: %Y. Matsushita, M. P. Gelfand and C. Ishii,
866: %J. Phys. Soc. Jpn. {\bf 66}, 3648 (1997);
867: %Z. Weihong,
868: %Phys. Rev. B {\bf 55}, 12267 (1997);
869: %R. R. P. Singh and N. Elstner,
870: %Phys. Rev. Lett. {\bf 81}, 4732 (1998).
871:
872: %%%%%% Haldane %%%%%%%%%%%%%%%%%
873: %\bibitem{Hald}
874: %A. Koga and N. Kawakami,
875: %Phys. Rev. B {\bf 61} 6133 (2000); preprint.
876:
877:
878: %%%%%% Neel boundary %%%%%%%%%%%%%%%%%
879: \bibitem{Neelb}
880: Y. Fukumoto and A. Oguchi,
881: J. Phys. Soc. Jpn. {\bf 67} (1998) 697;
882: J. Phys. Soc. Jpn. {\bf 67} (1998) 2205;
883: Z. Weihong, J. Oitmaa and C. J. Hamer,
884: Phys. Rev. B {\bf 58} (1998) 14147;
885: R. R. P. Singh, Z. Weihong, C. J. Hamer and J. Oitmaa,
886: Phys. Rev. B {\bf 60} (1999) 7278;
887: A. Koga, S. Kumada and N. Kawakami,
888: J. Phys. Soc. Jpn. {\bf 68} (1999) 2373; {\bf 69} (2000) 1843.
889:
890:
891: \end{thebibliography}
892:
893:
894: %\end{multicols}
895:
896: \end{document}
897:
898: