1: \documentclass[amsmath,amssymb,widetext,10pt,a4paper]{revtex4}
2: \usepackage{graphicx}
3:
4: \textheight 19.2truecm
5: \textwidth 11.3truecm
6: \evensidemargin 2.0truecm
7: \oddsidemargin 2.0truecm
8: \baselineskip 12pt
9: \topmargin 1.5truecm
10:
11: \begin{document}
12: \title{The resonating valence bond wave functions in quantum antiferromagnets}
13: \author{Alberto Parola}
14: \affiliation{Dipartimento di Fisica e Matematica, Universit\`a dell'Insubria, I-22100 Como, Italy}
15: \author{Sandro Sorella}
16: \author{Federico Becca}
17: \affiliation{INFM-Democritos, National Simulation Centre, and SISSA, I-34014 Trieste, Italy}
18: \author{Luca Capriotti}
19: \affiliation{Credit Suisse First Boston (Europe) Ltd. One Cabot Square, London E14 4QJ, United Kingdom}
20:
21: %--------------------------------------------------------------
22: %--------------------------------------------------------------
23:
24: \maketitle
25: Projected-BCS wave functions have been proposed as the paradigm
26: for the understanding of disordered spin states (spin liquids).
27: Here we investigate the properties of these wave functions showing how
28: Luttinger liquids, dimerized states, and gapped spin liquids may be described
29: by the same class of wave functions, which, therefore, represent an extremely
30: flexible variational tool. A close connection between spin liquids and
31: ``frozen'' superconductors emerges from this investigation.
32:
33: \vspace{15mm}
34: Many years after the first proposal~\cite{anderson},
35: the very existence of a spin-liquid ground state in two-dimensional (2D)
36: spin-$1/2$ models is still a controversial issue.
37: Short-range resonating-valence-bond (RVB)~\cite{figuerido} phases
38: with exponentially decaying spin-spin correlations
39: and no broken lattice symmetry (i.e., with no valence-bond order),
40: are conjectured to be stabilized by quantum fluctuations.
41: However, while it is possible to show that spin-liquid ground states
42: can be found in quantum dimer models~\cite{qdm}, the numerical evidence in
43: favor of disordered ground states in Heisenberg-like antiferromagnets
44: is still preliminary~\cite{rainbow,hex,kag2,mse}.
45: Renewed interest on this topic is triggered by the possible realization
46: of a frustrated Bose-Hubbard model with tunable interactions, by
47: trapping cold atomic clouds in optical lattices~\cite{tosi}.
48: A thorough analysis of the zero-temperature phase diagram of
49: these models will probably require an outstanding numerical effort,
50: however, it is possible to tackle the spin-liquid problem by a different,
51: less ambitious but remarkably informative, perspective.
52: Here, we present a detailed analysis of the properties of a class of RVB
53: wave functions, which have been shown to represent extremely good variational
54: states for a wide class of microscopic
55: Hamiltonians, ranging from effective low-energy models for correlated electrons
56: on the lattice to realistic models of atoms and
57: molecules~\cite{rainbow,dagotto,casula}. In particular, we will
58: show that both spin-liquid states and valence-bond crystals
59: may result from the ``freezing'' of Cooper pairs in the zero-doping limit
60: of correlated electron models.
61:
62: Following Anderson's suggestion~\cite{anderson}, we define the class of
63: projected-BCS (pBCS) wave functions in a $N$-site spin lattice,
64: starting from the ground state of a suitable BCS Hamiltonian:
65: \begin{eqnarray}
66: H(t,\Delta)&=&\sum_{i,j \sigma} (t_{ij}-\mu\,\delta_{ij})\, c^\dagger_{i,\sigma}c_{j,\sigma}
67: -\sum_{i,j} \left [ \Delta_{ij} c^\dagger_{i,\uparrow}c^\dagger_{j,\downarrow} +
68: \Delta_{ij}^* c_{j,\downarrow}c_{i,\uparrow} \right ]
69: \nonumber\\
70: &=&\sum_{k \sigma} (\epsilon_{k}-\mu)\, c^\dagger_{k,\sigma}c_{k,\sigma}
71: -\sum_{k} \left [ \Delta_{k} c^\dagger_{k,\uparrow}c^\dagger_{-k,\downarrow} +
72: \Delta_{k}^* c_{-k,\downarrow}c_{k,\uparrow} \right ],
73: \label{hbcs}
74: \end{eqnarray}
75: where the bare electron band $\epsilon_k$ is real and both $\epsilon_k$
76: and $\Delta_k$ are even functions of $k$. A chemical potential $\mu$ is
77: introduced in order to fix the number of electron equal to the number of sites.
78: In order to obtain a class of non-magnetic, translationally invariant,
79: singlet wave functions for spin-1/2 models, the ground state
80: $|BCS \rangle$ of Hamiltonian~(\ref{hbcs}) is then restricted to the physical
81: Hilbert space of singly-occupied sites by the Gutzwiller projector $P_G$.
82:
83: The first feature of such a wave function we want to discuss is the
84: {\it redundancy} implied by the electronic representation of a spin state,
85: i.e., the extra symmetries which appear when we write a spin state as
86: Gutzwiller projection of a fermionic state.
87: In turn, this property reflects in the presence of
88: a local, i.e., gauge, symmetry of the fermionic problem, as already
89: pointed out several years ago~\cite{su2}. Let us consider a generic
90: spin operator defined on a site in fermionic representation:
91: $X_{\alpha\beta}=c^\dag_\alpha c_\beta$.
92: This operator acts in the Hilbert subspace of singly-occupied sites (which
93: is left invariant under any spin Hamiltonian). It is easy to check that
94: the three SU(2) generators $N_0=(c^\dag_\uparrow c_\uparrow +
95: c^\dag_\downarrow c_\downarrow -1)/2$,
96: $N_+=c^\dag_\uparrow c^\dag_\downarrow$, $N_-=(N_+)^\dag$
97: commute, in the singly occupied site subspace, with $X_{\alpha\beta}$.
98: This property reflects the invariance of the operator $X_{\alpha\beta}$
99: under the usual $U(1)$ gauge transformation $g_\phi$:
100: \begin{equation}
101: g_\phi:\qquad c^\dag_\alpha \to e^{i\phi} c^\dag_\alpha,
102: \label{u1}
103: \end{equation}
104: and also by the SU(2) rotation $\Sigma_\theta$:
105: \begin{equation}
106: \Sigma_\theta: \qquad
107: \left (\begin{array}{c} c^\dag_\uparrow \cr c_\downarrow \end{array}\right )
108: \to \left ( \begin{array}{cc} \cos\theta & -i \sin\theta \cr
109: -i\sin\theta &\cos\theta \end{array} \right )
110: \left ( \begin{array}{c} c^\dag_\uparrow \cr c_\downarrow \end{array}\right ).
111: \label{sigma}
112: \end{equation}
113: These transformations are local, i.e., can be performed on each
114: site independently leaving every spin state invariant: they
115: generate the SU(2) gauge symmetry group.
116:
117: Let us now consider the Gutzwiller projected BCS state
118: in a Heisenberg-like model on a lattice with an even number $N=2n$ of sites.
119: \begin{equation}
120: |pBCS\rangle = P_G |BCS\rangle = P_G
121: \prod_k (u_k+v_k c^\dagger_{k,\uparrow}c^\dagger_{-k,\downarrow}) |0\rangle,
122: \label{bcs2}
123: \end{equation}
124: where the product is over all the $N$ wave vectors in the Brillouin zone.
125: The diagonalization of Hamiltonian~(\ref{hbcs}) gives explicitly
126: $$
127: u_k = \sqrt{{ E_k+\epsilon_k\over 2 E_k} } \qquad\qquad
128: v_k = {\Delta_k\over |\Delta_k |} \sqrt{{ E_k-
129: \epsilon_k\over 2 E_k} } \qquad\qquad
130: E_k = \sqrt{ \epsilon_k^2 +|\Delta_k|^2 } \qquad\qquad ,
131: $$
132: while the BCS pairing function $f_k$ is given by:
133: \begin{equation}
134: f_k = {v_k\over u_k} = {\Delta_k \over \epsilon_k + E_k}.
135: \label{fk}
136: \end{equation}
137: Clearly, the local gauge transformations previously defined
138: change the BCS Hamiltonian, breaking in general the translation
139: invariance. In the following, we will restrict to the class of
140: transformations which preserve the translational symmetry of the lattice
141: in the BCS Hamiltonian, i.e., the subgroup of {\it global} symmetries
142: corresponding to site independent angles $(\phi,\theta)$. By applying the
143: transformations~(\ref{u1}) and~(\ref{sigma}), the BCS Hamiltonian keeps
144: the same form with modified couplings:
145: \begin{eqnarray}
146: t_{ij}&\to& t_{ij} \nonumber\\
147: \Delta_{ij}&\to&\Delta_{ij} e^{2i\phi}
148: \label{phi}
149: \end{eqnarray}
150: for $g_\phi$, while the transformation $\Sigma_\theta$ gives:
151: \begin{eqnarray}
152: t_{ij}&\to& \cos 2\theta \, t_{ij} -i\sin \theta\cos\theta \, (\Delta_{ij} -\Delta^*_{ij}) \nonumber\\
153: &=&\cos 2\theta \, t_{ij}+\sin 2\theta \, {\rm Im}\Delta_{ij} \nonumber\\
154: \Delta_{ij}&\to& (\cos^2 \theta \, \Delta_{ij} + \sin^2\theta \, \Delta^*_{ij})- i\sin 2\theta \, t_{ij}
155: \nonumber\\
156: &=&{\rm Re} \Delta_{ij} + i\left ( \cos 2\theta \, {\rm Im}\Delta_{ij}-\sin 2\theta \, t_{ij} \right ).
157: \label{theta}
158: \end{eqnarray}
159: These relations are linear in $t_{ij}$ and $\Delta_{ij}$ and, therefore,
160: equally hold for the Fourier components $\epsilon_k$ and $\Delta_k$.
161: Note that, being $\Delta_r$ an even function, the real (imaginary) part of
162: its Fourier transform $\Delta_k$ equals the Fourier transform of the real
163: (imaginary) part of $\Delta_r$.
164: It is easy to see that these two transformations generate the full $O(3)$
165: rotation group on the vector whose components are
166: $(\epsilon_{k}, \,{\rm Re} \Delta_{k},\,{\rm Im}\Delta_{k})$.
167: As a consequence, the length $E_k$ of this vector is conserved by the full
168: group. In summary, this shows that there is an infinite number of different
169: translationally invariant BCS Hamiltonians which, after projection, give the
170: same spin state. Choosing a specific representation does not affect the
171: physics of the state but changes the pairing function $f_k$ before projection.
172: Within this class of states the only scalar under rotations, which can be
173: given some physical meaning, is the BCS energy spectrum $E_k$.
174: Clearly, the projection operator will modify the excitation spectrum
175: associated to the BCS wave function. Nevertheless, the invariance
176: with respect to SU(2) transformations suggests that $E_k$ may
177: reflect the nature of the physical excitation spectrum.
178:
179: Remarkably, in one dimension it is easy to prove that such a class of wave
180: functions is able to faithfully represent both the physics of Luttinger
181: liquids, appropriate for the nearest-neighbor Heisenberg model, and the
182: gapped spin-Peierls state, which is stabilized for sufficiently strong
183: frustration. In fact, it is known~\cite{haldane} that the simple choice
184: of nearest-neighbor hopping $t_{ij}$ ($\epsilon_k=-2t\,\cos k$, $\mu=0$ )
185: and vanishing gap function $\Delta_{ij}$ reproduces the exact solution of the
186: Haldane-Shastry model, while choosing a next-nearest neighbor
187: hopping ($\epsilon_k=-2t\,\cos 2k$, $\mu=0$) and a sizable nearest-neighbor
188: pairing ($\Delta=4\sqrt{2}t\,\cos k$) we recover the Majumdar-Gosh
189: state~\cite{maj}, i.e., the exact ground state of the frustrated Heisenberg
190: model when the next-nearest-neighbor exchange constant $J_2$ is half of the
191: nearest-neighbor one $J_1$. Note that in this case, the
192: BCS dispersion $E_k$ is strictly positive, i.e., the BCS Hamiltonian is gapped.
193:
194: A quantitative analysis of this class of wave functions can be only carried
195: out numerically. As an example, in Fig.1 we show that the spin-spin
196: correlations $G(r)$ have a long-range tail, i.e., $G(r) \sim 1/r$,
197: for both the known
198: $\Delta=0$ solution~\cite{haldane} but also for our pBCS state with
199: nearest-neighbor hopping and nearest- and third-neighbor $\Delta$,
200: whose energy is indeed remarkably accurate~\cite{dagotto}.
201: Moreover, Fig.2 shows that the same class of wave functions is able to
202: display a clear dimer ordering as soon as the BCS dispersion $E_k$ shows a gap
203: at the Fermi level. The agreement with the exact results (given by the
204: Density-Matrix Renormalization Group method) for the frustrated Heisenberg
205: model at $J_2/J_1=0.4$ is excellent also in this case.
206:
207: \begin{figure}[!ht]
208: \includegraphics[width=8cm]{fig1.eps}
209: \caption{Spin-spin correlation function at the largest distance:
210: $G(r)=(-1)^r\,\langle S_r^zS_1^z \rangle$ in a chain with L=2r sites as a
211: function of $r$.
212: Circles: exact results for the Heisenberg model, Squares: pBCS wave function
213: with nearest-neighbor hopping and third-neighbor $\Delta$.
214: Triangles: Gutzwiller projected Fermi sea.
215: A $1/r$ decay is clearly present in the variational wave functions.
216: The exact result shows logarithmic corrections as predicted by conformal
217: field theory.}
218: \end{figure}
219:
220: \begin{figure}[!ht]
221: \includegraphics[width=8cm]{fig2.eps}
222: \caption{Dimer-dimer correlation function
223: $\langle S_{x+1}^z S_x^z S_1^zS_0^z \rangle$
224: as a function of $x$ calculated via the best variational pBCS function
225: compared to the exact results obtained via Density-Matrix Renormalization
226: Group (DMRG) in a $150$-site chain.}
227: \end{figure}
228:
229: Now we specialize to the 2D square lattice and we
230: investigate whether it is possible to further
231: exploit the redundancy of the fermion representation of a
232: spin state in order to define a pairing function which,
233: before projection, breaks the reflection symmetries of the
234: lattice, while the projected state retains all the
235: correct quantum numbers. We will show that, if suitable conditions are
236: satisfied, a fully symmetric projected BCS state is
237: obtained from a BCS Hamiltonian with fewer symmetries than
238: the original spin problem.
239:
240: We first introduce a set of unitary operators.
241: \begin{itemize}
242: \item
243: {Spatial symmetries: $R_x (x,y)=(x,-y)$;
244: $R_{xy}(x,y)=(y,x)$. We define the transformation law of
245: creation operators $R\,c^\dagger_{j,\sigma}R^{-1}=c^\dagger_{Rj,\sigma}$
246: and the action of the operator on the vacuum $R\,|0\rangle=|0\rangle$.}
247: Note that these operators map each sublattice into itself.
248:
249: \item
250: { Particle-hole: $P_h c^\dagger_{j,\sigma} P_h^{-1}
251: = i\,(-1)^j c_{j,-\sigma}$, while the action of the $P_h$ operator
252: on the vacuum state is
253: $P_h|0\rangle=\prod_j c^\dagger_{j,\uparrow} c^\dagger_{j,\downarrow} |0\rangle$.}
254:
255: \item
256: { Gauge transformation: $G\,c^\dagger_{j,\sigma}G^{-1}=i\,c^\dagger_{j,\sigma}$
257: and $G\,|0\rangle=|0\rangle$.}
258:
259: \end{itemize}
260:
261: Clearly, $R_x$ and $R_{xy}$ are symmetries of the physical
262: problem (e.g., the Heisenberg model). $G$ is a symmetry because the
263: physical Hamiltonian has a definite number of electrons and
264: $P_h$ leaves invariant every configuration where each site is
265: singly occupied if the total magnetization vanishes: $N_\downarrow=
266: N_\uparrow=n$. Therefore, the previously defined operators commute
267: with the Heisenberg Hamiltonian and commute with each other
268: because reflections do not interchange the two sublattices.
269: The ground state of the Heisenberg model on a finite
270: lattice, if it is unique, must be simultaneous eigenstate
271: of all the symmetry operators. We want to investigate
272: sufficient conditions which guarantee that the projected
273: BCS state is indeed eigenstate of all these symmetries.
274:
275: Let us consider a hopping term which just couples the two
276: sublattices: $\epsilon_{k+Q}=-\epsilon_k$ and
277: a gap function $\Delta=\Delta_s+\Delta_{x^2-y^2}+
278: \Delta_{xy}$ with contributions from different symmetries.
279: Moreover, we consider the case in which $\Delta_s$ and $\Delta_{x^2-y^2}$
280: will couple different sublattices while $\Delta_{xy}$ is restricted to the
281: same sublattice. If this is the case, the BCS Hamiltonian
282: $H(t,\Delta_s,\Delta_{x^2-y^2},\Delta_{xy})$ transforms in the
283: following way under the different unitary operators:
284:
285: \begin{eqnarray}
286: R_x H(t,\Delta_s,\Delta_{x^2-y^2},\Delta_{xy}) R_x^{-1} &=&
287: H(t,\Delta_s,\Delta_{x^2-y^2},-\Delta_{xy}) \nonumber \\
288: R_{xy} H(t,\Delta_s,\Delta_{x^2-y^2},\Delta_{xy}) R_{xy}^{-1} &=&
289: H(t,\Delta_s,-\Delta_{x^2-y^2},\Delta_{xy}) \nonumber \\
290: P_h H(t,\Delta_s,\Delta_{x^2-y^2},\Delta_{xy}) P_h^{-1} &=&
291: H(t,\Delta_s^*,\Delta_{x^2-y^2}^*,-\Delta_{xy}^*) \nonumber \\
292: G H(t,\Delta_s,\Delta_{x^2-y^2},\Delta_{xy}) G^{-1} &=&
293: H(t,-\Delta_s,-\Delta_{x^2-y^2},-\Delta_{xy}). \nonumber
294: \end{eqnarray}
295:
296: Starting from these transformations, it is easy to define
297: suitable composite symmetry operators which indeed leave
298: the BCS Hamiltonian invariant. For instance,
299: let us consider the case in which $\Delta$ is real:
300: in this case we select the $R_xP_h$ and
301: $R_{xy}$ if $\Delta_{x^2-y^2}=0$ or $R_{xy}P_h G$ if $\Delta_s=0$.
302: We cannot set simultaneously $\Delta_{x^2-y^2}$ and $\Delta_s$
303: different form zero.
304: The eigenstates $|BCS\rangle$ of Eq.~(\ref{hbcs}) will be generally
305: simultaneous eigenstates of these two composite symmetry operators with
306: given quantum numbers, say $\alpha_x$ and $\alpha_{xy}$. Let us consider
307: the effect of projection over these states:
308: \begin{eqnarray}
309: \alpha_x\, P_G |BCS\rangle &=& P_G R_xP_h |BCS\rangle
310: = R_xP_h P_G |BCS\rangle =
311: \nonumber \\
312: &=& R_x P_G |BCS\rangle,
313: \label{eigenx}
314: \end{eqnarray}
315: where we have used that both $R_x$ and $P_h$ commute with the projector
316: and that $P_h$ acts as the identity on singly occupied states.
317: Analogously, when a $d_{x^2-y^2}$ gap is present,
318: \begin{eqnarray}
319: \alpha_{xy}\, P_G |BCS\rangle &=& P_G R_{xy}P_h G|BCS\rangle =
320: R_{xy}P_h G P_G |BCS\rangle = \nonumber \\
321: &=&(-1)^{n} R_{xy} P_G |BCS\rangle.
322: \label{eigenxy}
323: \end{eqnarray}
324:
325: These equations show that the projected BCS state with
326: both $xy$ and $x^2-y^2$ contributions to the gap has definite
327: symmetry under reflections, besides being translationally
328: invariant. The corresponding eigenvalues, for $n=N/2$ even,
329: coincide with the eigenvalues of the modified symmetry operators
330: $R_xP_h$ and $R_{xy}P_h G$ on the pure BCS state.
331: Note that the quantum numbers we have defined, refer to
332: the fermionic representation.
333: It turns out that an extremely good variational wave function
334: for the frustrated two dimensional Heisenberg antiferromagnet
335: can be obtained by including gap functions of different
336: symmetry~\cite{rainbow}.
337:
338: An alternative, but equally interesting representation of the
339: pBCS state can be given in terms of Slater determinants through the
340: following argument.
341: The pBCS wave function can be written in real space as:
342: \begin{equation}
343: |pBCS\rangle = P_G \left [ \sum_{R<X} \sum_{\sigma}
344: f(\sigma,R;-\sigma,X)
345: c^\dagger_{R,\sigma} c^\dagger_{X,-\sigma} \right ]^n |0\rangle
346: \label{bcs}
347: \end{equation}
348: where $R$ and $X$ run over the lattice sites and $\sigma$ is
349: the spin index. The antisymmetric pairing function
350: $f(\sigma,R;-\sigma,X)=-f(-\sigma,X;\sigma,R)$ is simply related to the
351: Fourier transform $f(r)$ of the previously defined $f_k$ of Eq.~(\ref{fk}):
352: $f(\uparrow,R;\downarrow,X)=f(R-X)$. In order to avoid double counting,
353: a given, arbitrary, ordering of the lattice sites has been assumed in
354: Eq.~(\ref{bcs}).
355: By expanding Eq.~(\ref{bcs}) and defining $f(\sigma,R;\sigma,X)=0$ we get
356: \begin{eqnarray}
357: |pBCS\rangle\,\, = \,\sum_{\begin{array}{c} R_1,\sigma_1\cdots R_n,\sigma_n \cr
358: X_1,\sigma_1^\prime \cdots X_n,\sigma_n^\prime \end{array}}
359: &f(\sigma_1,R_1;\sigma_1^\prime,X_1) \cdots f(\sigma_n,R_n;\sigma_n^\prime,X_n)&\,\,\times\nonumber \\
360: &|\sigma_1,R_1;\sigma_1^\prime,X_1\cdots \sigma_n,R_n;\sigma_n^\prime,X_n\rangle&
361: \label{mostro}
362: \end{eqnarray}
363: where the $2n$ labels ($R_1\cdots R_n;X_1\cdots X_n$) define a generic
364: partition of the $N=2n$ sites of the lattice.
365: We now rearrange the creation operators in the many body state of
366: Eq.~(\ref{mostro})
367: $$
368: |\sigma_1,R_1;\sigma_1^\prime,X_1\cdots \sigma_n,R_n;\sigma_n^\prime,X_n\rangle=
369: c^\dagger_{R_1,\sigma_1} c^\dagger_{X_1,\sigma_1^\prime}\cdots
370: c^\dagger_{R_n,\sigma_n} c^\dagger_{X_n,\sigma_n^\prime} |0\rangle,
371: $$
372: according to a given {\it site} ordering in the lattice (irrespective of the
373: spin).
374: This operation gives the ($S_z=0$) spin state $|\sigma_1\cdots \sigma_N\rangle$
375: multiplied by a phase factor equal to the sign $\epsilon$ of the associated
376: permutation. However, many distinct terms correspond to the same
377: spin state because they differ only by the way pairs are coupled:
378: $$
379: |pBCS\rangle = \sum_{\{\sigma_i\}}\sum_P \epsilon_P f(\sigma_1,R_1;\sigma_1^\prime,X_1)
380: \cdots f(\sigma_n,R_n;\sigma_n^\prime,X_n)
381: \,\,|\sigma_1\cdots \sigma_N\rangle
382: $$
383: Here $P$ runs over all the possible $(N-1)!!$ partitions of the $N=2n$ sites
384: of the lattices into pairs, for a given spin state
385: $|\sigma_1\cdots \sigma_N\rangle$ uniquely identified by the variables
386: $\{\sigma_i\}$.
387: The weight of the spin configuration
388: is exactly the Pfaffian of the $2n\times 2n$ antisymmetric matrix~\cite{pfaff}:
389: \begin{equation}
390: {\bf A} \,=\,\left ( \begin{array}{cc} \Big [ f(\uparrow,R_\alpha;\uparrow,R_\beta) \Big ] &
391: \Big [ f(\uparrow,R_\alpha;\downarrow,X_\beta)\Big ]\cr
392: \Big [ f(\downarrow,X_\alpha;\uparrow,R_\beta)\Big ] &
393: \Big [f(\downarrow,X_\alpha;\downarrow,X_\beta)\Big ]\end{array}\right ),
394: \label{matri}
395: \end{equation}
396: where the matrix has been written in terms of $n\times n$ blocks and $R_\alpha$
397: are the positions of the up spins in the $|\sigma_1\cdots \sigma_N\rangle$
398: state while $X_\alpha$ are the positions of the down spins.
399: The known relation
400: \begin{equation}
401: \left [ {\rm Pf} {\bf A} \right ]^2 = {\rm det} {\bf A},
402: \end{equation}
403: together with the fact that we considered $f(\sigma,R;\sigma,X)=0$
404: imply that ${\rm Pf} {\bf A} = {\rm det} \Big [ f(R_\alpha-X_\beta)\Big ]$~\cite{note}.
405: Therefore, the weight of the spin state $|\sigma_1\cdots \sigma_N\rangle$
406: may be written as the determinant of a matrix which depends on the BCS
407: pairing function.
408:
409: When we want to represent a spin state in a 2D lattice
410: by the pBCS wave function, special care must be given to the definition
411: of the phases of the pairing function $f(r)$.
412: As an example, let us consider a nearest-neighbor pairing function, i.e.,
413: we choose $|f(r)|=1$ when $r$ connects nearest neighbor sites in the lattice
414: and zero otherwise.
415: From Eq.~(\ref{bcs}) it is clear that the pBCS state is written as the sum
416: of all possible partitions of the $N$-site lattice into singlets
417: $(R_i,X_i)$ and the amplitude of a given partition is provided by the generic
418: term of the Pfaffian of the matrix ${\bf A}$:
419: \begin{equation}
420: \epsilon_P\,f(R_1-X_1) \cdots f(R_n-X_n).
421: \label{sign}
422: \end{equation}
423: Here we introduced the usual convention to orient a singlet from sublattice
424: $A$ to sublattice $B$ and we fix $R_i\in A$ ($X_i\in B$), while the
425: permutation $P$ relates the chosen ordering of sites in the lattice and the
426: sequence $(R_1,X_1\cdots R_n,X_n)$. Remarkably, as proved by
427: Kasteleyn~\cite{pfaff}, in {\it planar} lattices it is possible to choose
428: the phase of the function $f(r)$ so that the products of the form
429: Eq.~(\ref{sign}) have all the same sign. In such a case, the
430: pBCS wave function exactly reproduces the short-range RVB state:
431: the equal-amplitude superposition of all possible partitions of the lattice
432: into nearest-neighbor singlets~\cite{figuerido}.
433: In particular, on a rectangular lattice, the resulting
434: pairing function has $s+id$ symmetry~\cite{read}, i.e., $f(r)=1$ on
435: horizontal bonds and $f(r)=i$ on vertical bonds. In turn, this may be
436: obtained from the BCS Hamiltonian~(\ref{hbcs})
437: by considering third neighbor hopping
438: $\epsilon_k=-2t\,(\cos 2k_x + \cos 2k_y)$ (with $\mu=0$) and complex
439: nearest-neighbor gap function $\Delta_k=8t\,(\cos k_x + i\cos k_y)$.
440:
441: The properties of of the pBCS wave functions have been investigated by
442: Lanczos technique and variational Monte Carlo method in the frustrated
443: Heisenberg model on a square lattice, defined by the Hamiltonian:
444: \begin{equation}
445: H=J_1\sum_{\langle i,j\rangle} {\bf S}_i \cdot {\bf S}_j +
446: J_2\sum_{\langle \langle i,j\rangle \rangle} {\bf S}_i \cdot {\bf S}_j,
447: \label{j1j2}
448: \end{equation}
449: where the first sum runs over the nearest neighbors and the second on the
450: next-nearest neighbors (i.e., along the diagonal). When both coupling
451: constants are positive the model is frustrated.
452: In Fig.3 we compare the previously introduced dimer-dimer correlations
453: $\langle S_j^z S_i^z S_1^zS_0^z \rangle$
454: (where $(i,j)$ and $(0,1)$ are nearest neighbor sites) obtained via the
455: optimized pBCS wave function and the exact Lanczos results for the
456: Hamiltonian~(\ref{j1j2}) at $J_2/J_1=0.55$.
457: The figure shows that the pBCS wave function is able to capture the
458: correct behavior of correlations also in 2D.
459: In Fig.4 we report the size scaling of the squared dimer order parameter
460: obtained via the optimal variational wave function in the pBCS class.
461: The absence of dimer order is clearly suggested by the variational approach,
462: pointing toward the existence of a spin-liquid in the 2D $J_1{-}J_2$ model.
463:
464: \begin{figure}[!ht]
465: \includegraphics[width=8cm]{fig3.eps}
466: \caption{Dimer-dimer correlations in the $6\times 6$ square lattice for the
467: Hamiltonian~(\ref{j1j2}) at $J_2/J_1=0.55$. Stars: exact Lanczos results.
468: Triangles: optimized pBCS wave function.}
469: \end{figure}
470:
471: \begin{figure}[!ht]
472: \includegraphics[width=8cm]{fig4.eps}
473: \caption{Size scaling of the squared dimer order parameter as predicted by
474: the pBCS wave function.}
475: \end{figure}
476:
477: As a final remark, we like to comment on the fate of the insulating state
478: described by a pBCS wave function upon doping. When a limited number of
479: holes are injected into the lattice, it is likely that the basic structure
480: of the pBCS state is not affected by the presence of mobile charges,
481: being largely determined by the super-exchange interaction among spins.
482: As long as the BCS pairs present in the insulating state remain well
483: defined even at low doping, the system is generally expected to display
484: superconducting properties. This possibility has been
485: explicitly verified in the so-called $t{-}J$ model~\cite{dagotto},
486: where the natural generalization of a pBCS wave function has been shown
487: to give rise to off-diagonal long-range order.
488: Real materials, like high-temperature superconductors, display a considerably
489: richer physics and other effects may inhibit the actual realization of this
490: scenario: charge-density waves may occur at special values of the doping
491: or the gain in hole kinetic energy may induce a global change in the spin
492: state causing the breaking of electron
493: pairs. Only a detailed study of the microscopic Hamiltonian will discriminate
494: among these and other possibilities, and no general
495: statement can be drawn on the basis of our analysis. However,
496: our study suggests that a spin liquid (or even a valence bond crystal) may be
497: thought of as an insulating state adiabatically connected to a superconducting
498: phase which directly originates as soon as electrons are removed:
499: this is a remarkable manifestation of the effects of electron correlations
500: present in a gapped spin state which goes beyond standard band theory and the
501: Fermi Liquid approach.
502:
503: Partial support has been provided my MIUR through a PRIN grant. F.B. is
504: supported by INFM.
505:
506: %------ References --------------------
507:
508: \begin{thebibliography}{99}
509:
510: \bibitem{anderson} P.W. Anderson, Science {\bf 235}, 1196 (1987).
511:
512: \bibitem{figuerido} F. Figueirido, A. Karlhede, S. Kivelson, S. Sondhi,
513: M. Rocek, and D.S. Rokhsar, Phys. Rev. B {\bf 41}, 4619 (1989).
514:
515: \bibitem{qdm} D.S. Rokhsar and S.A. Kivelson, Phys. Rev. Lett. {\bf 61},
516: 2376 (1988); R. Moessner and S.L. Sondhi, Phys. Rev. Lett. {\bf 86}, 1881
517: (2001).
518:
519: \bibitem{rainbow} L. Capriotti, F. Becca, A. Parola, and S. Sorella,
520: Phys. Rev. Lett. {\bf 87}, 097201 (2001); Phys. Rev. B {\bf 67}, 212402 (2003).
521:
522: \bibitem{hex} J.B. Fouet, P. Sindzingre, and C. Lhuillier, Eur. Phys. J. B
523: {\bf 20}, 241 (2001).
524:
525: \bibitem{kag2} M. Mambrini and F. Mila, Eur. Phys. J. {\bf 17}, 651 (2001).
526:
527: \bibitem{mse} G. Misguich, C. Lhuillier, B. Bernu, and C. Waldtmann,
528: Phys. Rev. B {\bf 60}, 1064 (2000); W. Li Ming, G. Misguich, P. Sindzingre,
529: and C. Lhuillier, Phys. Rev. B {\bf 62}, 6372 (2000).
530:
531: \bibitem{tosi} M. Polini, R. Fazio, M.P. Tosi, J. Sinova, A.H. MacDonald,
532: Laser Physics, {\bf 14} 603 (2004).
533:
534: \bibitem{dagotto} S. Sorella, G.B. Martins, F. Becca, C. Gazza, L. Capriotti,
535: A. Parola, and E. Dagotto, Phys. Rev. Lett. {\bf 88}, 117002, (2002).
536:
537: \bibitem{casula} M. Casula and S. Sorella, J. Chem. Phys. {\bf 119}, 6500
538: (2003).
539:
540: \bibitem{su2} F.C. Zhang, C. Gros, T.M. Rice, and H. Shiba,
541: Supercond. Sci. Technol. {\bf 36}, 1 (1988).
542:
543: \bibitem{haldane} F.D.M. Haldane, Phys. Rev. Lett. {\bf 60}, 635 (1988).
544:
545: \bibitem{maj} C.K. Majumdar and D.K. Gosh, J. Math. Phys. {\bf 10}, 1388 (1969).
546:
547: \bibitem{pfaff} See for instance, P.W. Kasteleyn J. Math. Phys. {\bf 4},
548: 287 (1963).
549:
550: \bibitem{note} The ambiguity of the sign can be resolved by a careful
551: inspection.
552:
553: \bibitem{read} N. Read and B. Chakraborty, Phys. Rev. B {\bf 40}, 7133 (1989).
554:
555: \end{thebibliography}
556:
557: \end{document}
558: