cond-mat0011320/HF.tex
1: %\documentstyle[prb,aps,multicol,preprint,epsf]{revtex}
2: \documentstyle[prb,aps,multicol,epsf]{revtex}
3: 
4: 
5: \begin{document}
6: 
7: 
8: \draft
9: \title{Spin correlations in the two-leg antiferromagnetic ladder
10: in a magnetic field}
11: \author{T. Hikihara}
12: \address{Department of Earth and Space Science,
13: Graduate School of Science, Osaka University, \\
14: Toyonaka, Osaka 560-0043, Japan}
15: \author{A. Furusaki}
16: \address{Yukawa Institute for Theoretical Physics, Kyoto University,
17: Kyoto 606-8502, Japan}
18: \date{November 20, 2000}
19: \maketitle
20: \begin{abstract}
21: We study the ground-state spin correlations
22: in the gapless incommensurate regime of a $S=1/2$ $XXZ$ chain
23: and a two-leg antiferromagnetic ladder under a magnetic field,
24: in which the gapless excitations form a Tomonaga-Luttinger (TL) liquid.
25: We calculate numerically the two-spin correlation functions and
26: the local magnetization in the two models
27: using the density-matrix renormalization-group method.
28: By fitting the numerical results for an open $XXZ$ chain of 100
29: spins to correlation functions of a Gaussian model,
30: we determine the TL-liquid parameter $K$
31: and the amplitudes of the correlation functions.
32: The value of $K$ estimated from the fits is in excellent agreement
33: with the exact value obtained from the Bethe ansatz.
34: We apply the same method to the open ladder consisting of 200 spins
35: and determine the dependence of $K$ on the magnetization $M$.
36: The $K$-$M$ relation changes drastically
37: depending on the ratio of the coupling constants
38: in the leg and rung directions.
39: We also discuss implications of these results to experiments on
40: the nuclear spin relaxation rate $1/T_1$ and dynamical spin structure
41: factors.
42: 
43: 
44: 
45: 
46: \end{abstract}
47: \pacs{75.10.Jm,75.40.Cx,75.50.Ee,75.40.Mg}
48: 
49: 
50: 
51: 
52: \begin{multicols}{2}
53: 
54: 
55: 
56: 
57: \section{INTRODUCTION}
58: Spin ladder systems have been studied extensively over the past
59: decade.\cite{review}
60: There are reasons why the ladders have attracted so much attention.
61: Firstly, they naturally interpolate
62: one- and two-dimensional systems and may provide some hints
63: to better understand the high-temperature superconductivity
64: which occurs in square lattice ${\rm CuO_2}$ planes.
65: Secondly, spin ladders themselves have interesting physics and
66: deserve through investigation in their own right.
67: One of their most surprising properties is that low-energy physics
68: of spin ladders depends drastically on the number of legs.
69: Spin-$1/2$ antiferromagnetic (AF) ladders, for example,
70: have a finite gap in the spin excitation spectrum in the
71: even-leg case, whereas they have no gap in the odd-leg case.
72: The ground state of an even-leg ladder is a spin singlet and
73: its properties can be understood from the
74: short-range resonating-valence-bond picture.\cite{RVB}
75: This spin-gap behavior has been observed experimentally
76: on $S=1/2$ two-leg ladder compounds,\cite{exp1,exp2}
77: such as ${\rm SrCu_2O_3}$ and ${\rm Cu_2(C_5H_{12}N_2)_2Cl_4}$.
78: 
79: 
80: 
81: 
82: A gapless phase can appear in even-leg ladders when an
83: external field $h$ is applied.
84: If the field $h$ is larger than a critical field $h_{c1}$,
85: which is equal to the spin gap,
86: and if it is smaller than the saturation field $h_{c2}$,
87: then the ground state has a nonzero magnetization $M$ and the energy gap
88: between the ground state and the first-excited states vanishes.
89: The gapless mode has been shown to be described as a
90: Tomonaga-Luttinger (TL) liquid both in the strong- and weak-coupling
91: limit,\cite{Chi-Gia,Gia-Tsv,Furu-Zhn}
92: where the coupling in the rung-direction $J_\perp$ is much larger or
93: much smaller than the one in the leg-direction $J_\parallel$,
94: respectively.
95: However, the $M$ dependence of the TL-liquid parameter $K$,
96: which governs the spin correlations in long wave length,
97: has been obtained analytically only in the strong-coupling limit
98: and it remains as a nontrivial problem to determine $K$ for general
99: $J_\perp/J_\parallel$.
100: In the gapless phase the system shows incommensurate spin correlations
101: since the Fermi wavenumber of Jordan-Wigner fermions is shifted
102: from $\pi/2$ in the presence of the magnetic field
103: which acts as a chemical potential for the fermions.
104: The wavenumber $Q$ characterizing the incommensurability of the
105: gapless mode varies continuously as $h$ increases.
106: This gapless incommensurate (IC) phase is in fact in the same
107: universality class as the one-dimensional $S = 1/2$ $XXZ$ model
108: in a magnetic field, as we will see.
109: 
110: 
111: 
112: 
113: In this paper, we study low-energy properties of
114: the $S = 1/2$ two-leg AF ladder in the gapless IC regime
115: for broad range of $J_\perp / J_\parallel$.
116: We show that the system is a TL liquid for arbitrary
117: $J_\perp / J_\parallel$
118: and determine the $M$ dependence of $K$ numerically.
119: To this end, we compute numerically the ground-state spin-correlation
120: functions and the local magnetization in the open ladders
121: using the density-matrix renormalization-group (DMRG)
122: method~\cite{White1,White2}
123: and extract the TL-liquid parameter by fitting the data to
124: correlation functions obtained from the Abelian bosonization.
125: This method was applied in our previous work~\cite{CorAm}
126: to the $S=1/2$ $XXZ$ chain at $h = 0$ and proved to be effective
127: in determining both the TL-liquid parameter and amplitudes of
128: correlation functions.
129: In order to demonstrate the validity of the analysis in the gapless IC
130: phase, we first apply it to the $S = 1/2$ $XXZ$ chain for $h > 0$.
131: The model is exactly solvable by the Bethe ansatz
132: and the TL-liquid parameter $K$ can be calculated
133: for arbitrary value of $M$.
134: It thus provides a good test ground to check
135: accuracy of our method.
136: We find that $K$ estimated from the DMRG data is in excellent
137: agreement with the exact calculation.
138: We then apply the same method to the two-leg ladders in a magnetic
139: field.
140: Our numerical data of correlation functions are fitted well for
141: broad range of $J_\perp/J_\parallel$ to the formulas based on the
142: bosonization approach, confirming that the gapless modes are in fact
143: in the universality class of a TL liquid for arbitrary
144: $J_\perp/J_\parallel$.
145: The $M$ dependence of $K$ obtained in the large $J_\perp/J_\parallel$
146: limit agrees with the analytic result obtained through mapping to the
147: $XXZ$ chain.
148: In this limit $K$ is less than 1 for $0 < M < 1$.
149: As $J_\perp/J_\parallel$ decreases,
150: $K$ increases and become larger than 1 for intermediate values of $M$.
151: Our numerical result indicates that $K$ takes a universal value 1 for any
152: $J_\perp/J_\parallel$ in the limits $M\to0$ and $M\to1$.
153: 
154: 
155: 
156: 
157: The plan of the paper is as follows.
158: We first review the Abelian bosonization approach
159: to the $S=1/2$ $XXZ$ chain under a magnetic field in Sec.~II A.
160: The formulas of the spin correlations and the local magnetization
161: in finite open chains are presented.
162: In Sec.~II B, we show numerical data for the $XXZ$ chain of $L=100$
163: sites obtained from the DMRG calculation and fit the data to the
164: functions given in Sec.~II A.
165: In Sec.\ III A, we briefly review some relevant results of the previous
166: analytic studies on the two-leg ladders
167: in the strong- and weak-coupling limits.
168: The DMRG data and the results of fitting on the open ladders
169: with $L = 200$ sites are shown in Sec.\ III B.
170: The $M$ dependence of $K$ for various values of $J_\perp/J_\parallel$
171: is obtained.
172: Its implications to NMR and neutron scattering experiments are briefly
173: mentioned.
174: Finally, our results are summarized in Sec.\ IV.
175: 
176: 
177: 
178: 
179: 
180: \section{$XXZ$ CHAIN}
181: \subsection{Bosonization approach}
182: In this section, we consider spin-1/2 $XXZ$ chains with open ends in a
183: magnetic field $h$.
184: The Hamiltonian is
185: \begin{equation}
186: {\cal H}_{\rm ch} = {\cal H}_0 + {\cal H}_h \label{eq:Hchn}
187: \end{equation}
188: with
189: \begin{eqnarray}
190: {\cal H}_0 &=&
191:    J \sum_{l=1}^{L-1} (\bbox{S}_l, \bbox{S}_{l+1} )_\Delta,
192:     \nonumber \\
193: {\cal H}_h &=& - h \sum_{l=1}^L S^z_l, \nonumber
194: \end{eqnarray}
195: where $\bbox{S}_l$ are $S = 1/2$ spin operators and
196: $(\bbox{S}_l, \bbox{S}_{l'} )_\Delta =
197:    S^x_l S^x_{l'} + S^y_l S^y_{l'} + \Delta S^z_l S^z_{l'}$.
198: We assume the system size $L$ to be even throughout this paper
199: and treat only the case where $J > 0$ and $0 \le \Delta \le 1$.
200: We note that the Hamiltonian (\ref{eq:Hchn}) for $-1 < \Delta\le 1$
201: can be solved exactly by Bethe ansatz
202: for arbitrary values of $h$.\cite{Bethe0,Bethe1,Bethe2}
203: 
204: 
205: 
206: 
207: We use the standard Abelian bosonization techniques to analyze
208: spin-spin correlation functions at zero temperature.
209: We basically follow the scheme presented in
210: Ref.~\onlinecite{Eggert} and generalize it to the case of open chains
211: in magnetic fields.
212: The bosonization formulas in the absence of magnetic fields are
213: reported in Ref.~\onlinecite{CorAm}.
214: 
215: 
216: 
217: 
218: The low-energy dynamics of $XXZ$ chains is described
219: by the Gaussian model,\cite{Bethe1}
220: \begin{equation}
221: \widetilde{{\cal H}}_{\rm ch} =
222:    \frac{v}{2} \int_0^{L+1} dx
223:    \left[ \left( \frac{d\phi}{dx}\right)^2
224:          + \left( \frac{d\tilde{\phi}}{dx}\right)^2 \right] ,
225:            \label{eq:HchnBos}
226: \end{equation}
227: where $v$ is the spin-wave velocity.
228: The continuous variable $x$ is identified with the site index $l$
229: under the assumption that the lattice spacing equals unity.
230: The bosonic fields $\phi(x)$ and $\tilde{\phi}(x)$ obey
231: the commutation relation
232: $[\phi(x), \tilde{\phi}(y)] = -i \Theta(x-y)$,
233: where $\Theta(x)$ is the step function.
234: The spin operators in the original Hamiltonian (\ref{eq:Hchn}) are
235: related to the bosonic fields by the relations,
236: \begin{eqnarray}
237: S^z_l &=&
238: \frac{1}{2 \pi R}\frac{d\phi}{dx}
239: + a (-1)^l \sin\left(\frac{\phi(l)}{R}\right),
240:            \label{eq:Szchn} \\
241: S^{-}_l &=&
242: \exp[- i 2\pi R\tilde{\phi}(l)]
243: \left[ b \sin\left(\frac{\phi(l)}{R}\right) + c(-1)^l \right],
244:       \label{eq:S-chn}
245: \end{eqnarray}
246: with $a$, $b$, and $c$ being real constants.
247: The parameter $R$ determines the exponents of correlation functions.
248: We also introduce the TL-liquid parameter $K$
249: by $K = 1/(4\pi R^2)$;
250: $K=1$ in the $XY$ case ($\Delta=0$), and
251: $K=1/2$ in the Heisenberg case ($\Delta=1$) at $h=0$.
252:   From Eq.~(\ref{eq:S-chn}), $S^x_l$ is written as
253: \begin{equation}
254: S^x_l = c(-1)^l \cos[2\pi R\tilde{\phi}(l)]
255:          - i b \sin[2\pi R\tilde{\phi}(l)]
256:            \sin\left(\frac{\phi(l)}{R} \right).
257:      \label{eq:Sxchn}
258: \end{equation}
259: The second term with the coefficient $ib$ is Hermitian due to the
260: commutation relation $[\phi(l),\tilde\phi(l)]=-i/2$.
261: The open boundary conditions are translated to
262: the boundary conditions on the bosonic fields at the two
263: phantom sites $l=0$ and $l=L+1$:\cite{Eggert}
264: $\phi(0)=0$ and $\phi(L+1)=2\pi RLM_{\rm ch}$, where $M_{\rm ch}$ is
265: the magnetization per site,
266: \begin{equation}
267: M_{\rm ch} = \frac{1}{L} \sum_{l=1}^L S^z_l.
268: \label{eq:Mch}
269: \end{equation}
270: The total magnetization $LM_{\rm ch}$ is an integer for even $L$.
271: These boundary conditions lead to the mode expansion,
272: \begin{eqnarray}
273: \phi(x) &=&
274:    \frac{x}{L+1} \phi_0
275:    + \sum_{n=1}^\infty \frac{\sin(q_n x)}{\sqrt{\pi n}}
276:           \left( a_n + a_n^\dagger \right) , \label{eq:mode1} \\
277: \tilde{\phi}(x) &=&
278:    \tilde{\phi}_0
279:    + i \sum_{n=1}^\infty \frac{\cos(q_n x)}{\sqrt{\pi n}}
280:          \left( a_n - a_n^\dagger \right) ,   \label{eq:mode2}
281: \end{eqnarray}
282: where $q_n = \pi n /(L+1)$, $[\tilde{\phi}_0, \phi_0] = i$,
283: and $a_m$ are boson annihilation operators obeying
284: $[a_m, a_n^\dagger] = \delta_{m,n}$.
285: Note that the commutation relation between $\phi(x)$ and
286: $\tilde{\phi}(y)$ mentioned above is satisfied.
287: The lowest energy state $|M_{\rm ch}\rangle$ in the subspace
288: in which the magnetization per spin is $M_{\rm ch}$ is a vacuum of
289: $a_n$
290: \begin{equation}
291: a_n|M_{\rm ch}\rangle = 0
292: \end{equation}
293: and an eigenstate of $\phi_0$
294: \begin{equation}
295: \phi_0 |M_{\rm ch}\rangle = 2\pi R L M_{\rm ch}|M_{\rm ch}\rangle.
296: \label{eq:phi_0}
297: \end{equation}
298: We may regard $\tilde\phi_0$ as a coordinate variable along a
299: fictitious ring of radius $1/2\pi R$ and take
300: $\phi_0=-id/d\tilde\phi_0$ to be its momentum conjugate.
301: The state $|M_{\rm ch}\rangle$ is then proportional to
302: $\exp(i2\pi RLM_{\rm ch}\tilde\phi_0)$.
303: The bosonization formulas (\ref{eq:Szchn}) and
304: (\ref{eq:S-chn}) represent only the leading contributions.
305: In the next order $S^-_l$ has a term of the form
306: $(-1)^l\exp[-2\pi iR\tilde\phi(l)]\cos[2\phi(l)/R]$.
307: We will, however, ignore this contribution because it yields only
308: subleading corrections that disappear quickly for large $|l-l'|$.
309: 
310: 
311: 
312: 
313: Using Eqs.~(\ref{eq:Szchn})--(\ref{eq:phi_0}), one can evaluate the
314: two-spin correlation functions
315: $\langle S^\alpha_l S^\alpha_{l'} \rangle$ $(\alpha = x, z)$ and
316: the local magnetization $\langle S^z_l \rangle$ in open chains,
317: where $\langle \cdots \rangle$ denotes the expectation value
318: in the state $|M_{\rm ch}\rangle$.
319: Brief account of their derivation is given in Appendix.
320: Here we present the final results:
321: \end{multicols}
322: \begin{eqnarray}
323: \langle S^x_l S^x_{l'} \rangle &\equiv&
324: X(l,l';q) \nonumber\\
325: &=&
326:       \frac{f_{\eta/2}(2l) f_{\eta/2}(2l')}{f_\eta(l-l') f_\eta(l+l')}
327:      \left[ \frac{c^2}{2} (-1)^{l-l'}
328: % \right. \nonumber  \\ &&
329:     + \frac{bc}{2} {\rm sgn}(l-l') \left(
330:       \frac{(-1)^l \cos(q l')}{f_{1/2\eta}(2l')}
331:     - \frac{(-1)^{l'} \cos(q l)}{f_{1/2\eta}(2l)} \right)
332:      \right. \nonumber \\
333: && \left.\hspace*{3cm}
334:     - \frac{b^2}{4 f_{1/2\eta}(2l) f_{1/2\eta}(2l')}
335:         \left( \cos\left[ q (l+l')\right]
336:                   \frac{f_{1/\eta}(l-l')}{f_{1/\eta}(l+l')}
337: %   \right. \nonumber \\ && \left.
338:               + \cos\left[ q (l-l')\right]
339:                 \frac{f_{1/\eta}(l+l')}{f_{1/\eta}(l-l')} \right)
340:    \right],
341:                          \label{eq:Cxechn} \\
342: \langle S^z_l S^z_{l'} \rangle &\equiv&
343: Z(l,l';q) \nonumber\\
344: &=&
345:      \frac{q}{2\pi}
346:      \left(\frac{q}{2\pi}
347:             + a \frac{(-1)^l \sin(q l)}{f_{1/2\eta}(2l)}
348:             + a \frac{(-1)^{l'} \sin(q l')}{f_{1/2\eta}(2l')}
349:      \right)
350: %          \nonumber \\ &&
351:     - \frac{1}{4\pi^2 \eta} \left(
352:             \frac{1}{f_2(l-l')} + \frac{1}{f_2(l+l')} \right)
353:             \nonumber \\
354:     && + \frac{a^2}{2}\frac{(-1)^{l+l'}}{f_{1/2\eta}(2l)f_{1/2\eta}(2l')}
355:           \left( \cos\left[q (l-l')\right]
356:                        \frac{f_{1/\eta}(l+l')}{f_{1/\eta}(l-l')}
357: %      \right. \nonumber\\ &&
358:                - \cos\left[q (l+l')\right]
359:                   \frac{f_{1/\eta}(l-l')}{f_{1/\eta}(l+l')} \right)
360:              \nonumber \\
361:     && + \frac{a}{2\pi \eta} \left(
362:           \frac{(-1)^l \cos(q l)}{f_{1/2\eta}(2l)}
363:                    [g(l+l')+g(l-l')]
364: %   \right.  \nonumber \\  & & \left.~~~~~~~~
365:       + \frac{(-1)^{l'} \cos(q l')}{f_{1/2\eta}(2l')}
366:                    [g(l+l')-g(l-l')]  \right) ,
367:       \label{eq:Czechn} \\
368: \langle S^z_l \rangle &\equiv&
369: z(l;q)=
370: \frac{q}{2\pi}
371:    + a \frac{(-1)^l \sin(q l)}{f_{1/2\eta}(2l)},
372:          \label{eq:Szechn}
373: \end{eqnarray}
374: \begin{multicols}{2}
375: where
376: \begin{eqnarray}
377: &&
378: \eta=2\pi R^2=\frac{1}{2K},\\
379: &&
380: f_\alpha(x) =
381:    \left[
382: \frac{2(L+1)}{\pi}\sin\left(\frac{\pi |x|}{2(L+1)} \right)
383: \right]^\alpha,
384:    \label{eq:fx} \\
385: &&
386: g(x) = \frac{\pi}{2(L+1)} \cot\left( \frac{\pi x}{2(L+1)}\right).
387:           \label{eq:gx}
388: \end{eqnarray}
389: The wavenumber $q$, characterizing the IC character of
390: the spin correlations in a magnetic field, is related to $M_{\rm ch}$
391: by
392: \begin{equation}
393: q = \frac{2\pi M_{\rm ch}L}{L+1}.
394: \label{eq:Qch}
395: \end{equation}
396: The factor $L/(L+1)$ appears as a
397: result of the open boundary conditions.
398: Under the periodic boundary conditions $q$ should be simply
399: equal to $2\pi M_{\rm ch}$, because the first term in
400: Eq.~(\ref{eq:mode1}) is $\phi_0 x/L$ in this case.
401: This term must be $\phi_0 x/(L+1)$ in the open-boundary case in order
402: for $\phi(x)$ and $\tilde\phi(x)$ to satisfy the commutation relation
403: in the interval $[0,L+1]$.
404: We emphasize that Eqs.~(\ref{eq:Czechn}) and (\ref{eq:Szechn})
405: reproduce the exact results for the $XY$ chain when $\eta=1/2$ and
406: $a=-1/\pi$.
407: As is well known, the $XXZ$ spin chain is equivalent to a model of
408: spinless fermions with nearest-neighbor interaction.
409: In this model, $S^z_l$ is none but the fermion density, and
410: the oscillating term in Eq.~(\ref{eq:Szechn}) corresponds to the
411: Friedel oscillations near the open
412: ends.\cite{Friedel1,Friedel2,Friedel3,Friedel4}
413: 
414: 
415: 
416: 
417: In the thermodynamic limit ($L \to \infty$) with $|l-L/2| \ll L$ and
418: $|l'-L/2| \ll L$, the spin correlations have the asymptotic forms
419: \begin{eqnarray}
420: \langle S^x_l S^x_{l'} \rangle &=&
421:    A_x\frac{(-1)^{l-l'}}{|l-l'|^\eta}
422:    - \widehat{A}_x \frac{\cos\left[q (l-l')\right]}
423:                         {|l-l'|^{\eta+1/\eta}},
424:             \label{eq:CxLchn} \\
425: \langle S^z_l S^z_{l'} \rangle &=& {M_{\rm ch}}^2
426:    + A_z (-1)^{l-l'} \frac{\cos\left[q (l-l')\right]}
427:                          {|l-l'|^{1/\eta}}  \nonumber \\
428: &&- \frac{1}{4\pi^2 \eta |l-l'|^2},
429:             \label{eq:CzLchn}
430: \end{eqnarray}
431: where the correlation amplitudes $A_x$, ${\widehat{A}_x}$, and $A_z$
432: are related to the numerical constants
433: $a$, $b$, and $c$ by $A_x = c^2/2$,
434: ${\widehat{A}_x} = b^2/4$, and $A_z = a^2/2$.
435: We can therefore estimate the TL-liquid parameter and
436: the correlation amplitudes in the thermodynamic limit
437: by extracting the fitting parameters $R$, $a$, $b$, and $c$
438: from the numerical data on a finite system with use of
439: Eqs.~(\ref{eq:Cxechn}), (\ref{eq:Czechn}), and (\ref{eq:Szechn}).
440: At the same time, the TL-liquid parameter $K$ can be calculated exactly for
441: any $M_{\rm ch}$ by solving an integral equation obtained from the
442: Bethe ansatz.\cite{TLpr1,TLpr2}
443: We will compare our estimates of $K$ obtained from the fitting procedure
444: with the Bethe ansatz results in the next subsection.
445: 
446: 
447: 
448: 
449: \subsection{Numerical results}
450: Using the DMRG method,\cite{White1,White2}
451: we computed the two-spin correlation functions
452: $\langle S^\alpha_l S^\alpha_{l'} \rangle$ $(\alpha=x,z)$ and
453: the local magnetization $\langle S^z_l \rangle$ in the $L=100$ open
454: chains.
455: The two-point functions were calculated for $l = r_0 - r/2$ and
456: $l' = r_0 + r/2$, where $r_0 = L/2$ for even $r$ and $r_0 = (L+1)/2$
457: for odd $r$.
458: The calculation was performed for each lowest-energy state of
459: ${\cal H}_0$ in the subspace of various values of $M_{\rm ch}$.
460: We employed the finite system algorithm of improved
461: version.\cite{White3}
462: The maximum number of kept states $m$ is $100$.
463: We estimate the numerical error due to the truncation of the Hilbert
464: space from the difference between the data with $m = 100$ and those
465: with $m = 70$.
466: The estimated errors for $\langle S^x_l S^x_{l'} \rangle$,
467: $\langle S^z_l S^z_{l'} \rangle$, and $\langle S^z_l \rangle$ are,
468: at largest, of order $10^{-5}$,$10^{-6}$, and $10^{-6}$, respectively.
469: 
470: 
471: 
472: 
473: In Fig.~\ref{fig:chn}, we show the spin correlations
474: $\langle S_l^\alpha S_{l'}^\alpha \rangle$
475: ($\alpha = x,z$) and the local magnetization $\langle S_l^z \rangle$
476: at $\Delta = 0.5$ for three different values of $M_{\rm ch}$.
477: The DMRG data are shown by open symbols whose sizes
478: are larger than the truncation error mentioned above.
479: Taking $R$, $a$, $b$, and $c$ as fitting parameters, we fit
480: the numerical data to Eqs.~(\ref{eq:Cxechn})--(\ref{eq:Szechn}).
481: The results of the fitting using the DMRG data of
482: $\langle S_l^\alpha S_{l'}^\alpha \rangle$
483: for $10 \le r \le 90$ and of $\langle S_l^z \rangle$ for
484: $1 \le l \le 100$ are also plotted in the figure by the small solid
485: symbols.
486: One can see that the fits are in excellent agreement with
487: the DMRG data, proving the validity of the bosonization formulas
488: (\ref{eq:Cxechn})--(\ref{eq:Szechn}).
489: 
490: 
491: 
492: 
493: For various values of $M_{\rm ch}$ and $\Delta$, we determined the
494: parameters $R$, $a$, $b$, and $c$ by the fitting procedure.
495: In doing so we used numerical data of several ranges,
496: $10 \le r \le 80$, $10 \le r \le 90$,
497: $20 \le r \le 80$, and $20 \le r \le 90$ for the two-spin correlation
498: functions and $1 \le l \le 100$ and $10 \le l \le 90$ for
499: the local magnetization.
500: We take the mean and the variance of the fitting parameters obtained
501: for the different ranges of $r$ and $l$
502: as the estimated value and the error of the estimates,
503: respectively.
504: The TL-liquid parameter $K \equiv 1/(4\pi R^2)$ estimated from the
505: numerical data of $\langle S^x_l S^x_{l'} \rangle$
506: is plotted as a function of $M_{\rm ch}$ in Fig.~\ref{fig:Kchn}.
507: The exact values obtained from the Bethe ansatz
508: method~\cite{TLpr1,TLpr2} are also shown as dotted lines.
509: The agreement is excellent.
510: We also estimated $K$ from the fitting of $\langle S^z_l S^z_{l'}\rangle$
511: and $\langle S^z_l\rangle$ and obtained similar results as
512: Fig.~\ref{fig:Kchn}.
513: We found, however, that the estimates from the last two correlators
514: show some deviations from the Bethe ansatz results when $K$
515: is small.
516: We do not exactly know why they deviate.
517: One possible reason might be the effect of the leading irrelevant
518: operator neglected in the
519: Gaussian model that becomes marginal at $K=1/2$.
520: In the $XY$ regime of our interest,
521: spins have stronger correlations in the $S^x$ and $S^y$ components
522: than in $S^z$, and thus we may expect that
523: $\langle S^x_l S^x_{l'}\rangle$ should give us most reliable estimates.
524: 
525: 
526: 
527: 
528: 
529: For the correlation amplitudes $A_x$ and $A_z$,
530: Lukyanov and Zamolodchikov conjectured the exact formulas which are
531: valid at $h=0$,\cite{Lu-Za,Luky}
532: \end{multicols}
533: \begin{eqnarray}
534: A_x^{\rm LZ} &=&
535: \frac{1}{8(1-\eta)^2}
536: \left[\frac{\Gamma(\frac{\eta}{2(1-\eta)})}
537:              {2\sqrt{\pi}\,\Gamma(\frac{1}{2(1-\eta)})}
538: \right]^\eta
539: % \nonumber \\ &&\times
540: \exp\left[
541: -\int^\infty_0\frac{dt}{t}
542:    \left(\frac{\sinh(\eta t)}{\sinh(t)\cosh[(1-\eta)t]}
543:          -\eta e^{-2t}\right)\right],
544: \label{eq:LZx} \\
545: A_z^{\rm L} &=&
546: \frac{2}{\pi^2}
547: \left[\frac{\Gamma(\frac{\eta}{2(1-\eta)})}
548:              {2\sqrt{\pi}\,\Gamma(\frac{1}{2(1-\eta)})}
549: \right]^{1/\eta}
550: % \nonumber \\ &&\times
551: \exp\left[
552:    \int^\infty_0\frac{dt}{t}
553:    \left(\frac{\sinh[(2\eta-1) t]}{\sinh(\eta t)\cosh[(1-\eta)t]}
554:          -\frac{2\eta -1}{\eta} e^{-2t}\right)\right],
555: \label{eq:LZz}
556: \end{eqnarray}
557: \begin{multicols}{2}\noindent
558: where $\Gamma(x)$ is the Gamma function.
559: Equations (\ref{eq:LZx}) and (\ref{eq:LZz}) have been confirmed
560: numerically.\cite{CorAm,Luky}
561: In Table~\ref{tab:Achn}, we give our estimates of
562: the correlation amplitudes
563: $A_x = c^2/2$ and $A_z = a^2/2$ obtained from
564: the fitting of $\langle S^x_l S^x_{l'}\rangle$
565: and $\langle S^z_l \rangle$ for $0 < M_{\rm ch} < 0.5$,
566: together with the exact values (\ref{eq:LZx}) and (\ref{eq:LZz}) at
567: $M_{\rm ch} = 0$.~\cite{no hat Ax}
568: As can be seen in Table~\ref{tab:Achn} (a),
569: $A_x$ decreases monotonically from the value given by
570: Eq.~(\ref{eq:LZx}) to zero as $M_{\rm ch}$ increases from 0 to $1/2$.
571: Thus, $A_x$ depends not only on $K$ but also on $M_{\rm ch}$.
572: (See, for example, the data for $\Delta=0$ where $K$
573: takes a constant value 1 for any $M_{\rm ch}$.)
574: When $M_{\rm ch}$ approaches $1/2$, where $K\to1$, $A_x$ seems to go
575: to zero linearly for any $\Delta$.
576: This can be easily understood once we consider one-magnon contribution
577: to the correlation function.
578: On the other hand, $A_z$ decreases monotonically from
579: the number given by Eq.~(\ref{eq:LZz})
580: to the universal value $A_z = 1/(2\pi^2) \simeq 0.05066$
581: as $M_{\rm ch}$ increases from 0 to $1/2$.~\cite{Ampnote}
582: An exception is the case $\Delta = 0$,
583: where $A_z = 1/(2\pi^2)$ for any $M_{\rm ch}$.
584: The convergence of $A_z$ to $1/(2\pi^2)$ at $M_{\rm ch} \to 1/2$
585: is consistent with the fact that
586: the correlator $\langle S^z_l S^z_{l'}\rangle$
587: must take a constant value $1/4$ at $M_{\rm ch}=1/2$.
588: The right-hand side of Eq. (\ref{eq:CzLchn}) equals
589: ${M_{\rm ch}}^2$ when $\eta = 1/2$, $q = \pi$, and $A_z = 1/(2\pi^2)$.
590: 
591: 
592: 
593: \section{TWO-LEG AF LADDER}
594: Encouraged by the success in the last section, we study the two-leg AF
595: ladders in a magnetic field using the same method.
596: We begin with a brief review of the analytic results on the ladder
597: in the strong- and weak-coupling limits.
598: 
599: 
600: 
601: 
602: \subsection{Review of Analytic Results}
603: The Hamiltonian of the open two-leg ladder studied in this section is
604: given by
605: \begin{eqnarray}
606: {\cal H} & = &
607:    J_\parallel \sum_{\mu = 1,2} \sum_{l=1}^{L-1}
608:     (\bbox{S}_{\mu,l},\bbox{S}_{\mu,l+1} )_\Delta
609:    + J_\perp \sum_{l=1}^L (\bbox{S}_{1,l},\bbox{S}_{2,l} )_\Delta
610:    \nonumber \\ &&
611:    - h \sum_{\mu=1,2} \sum_{l=1}^L S_{\mu,l}^z.  \label{eq:Hlad}
612: \end{eqnarray}
613: The anisotropy $\Delta$ is introduced for generality.
614: We assume that the coupling in the leg- and rung-direction,
615: $J_\parallel$ and $J_\perp$, are positive (antiferromagnetic).
616: The spin ladder has an excitation gap in weak magnetic fields
617: $h < h_{c1}$.
618: We concentrate on the ladder in the gapless IC regime, i.e., in
619: the case $h_{c1} \le h \le h_{c2}$.
620: We denote the ratio $J_\perp / J_\parallel$ by $j$ hereafter.
621: 
622: 
623: 
624: 
625: We begin with the strong-coupling limit ($j \gg 1$), for which
626: a simple intuitive picture is available.
627: It is known that the system in this limit can be mapped to an
628: effective $S=1/2$ $XXZ$ chain,\cite{Furu-Zhn,Mila,Totsu} as we
629: explain below.
630: Let us first assume $J_\parallel = 0$.
631: In this case, an eigenstate of ${\cal H}$ is written as
632: a direct product of rung states.
633: At each rung two spins $\bbox{S}_{1,l}$
634: and $\bbox{S}_{2,l}$ are either in a singlet state
635: $|s_l \rangle = ( |\uparrow \downarrow \rangle
636:                    - |\downarrow \uparrow \rangle ) / \sqrt{2}$
637: or in one of the triplet states,
638: $|t_l^+ \rangle = |\uparrow \uparrow \rangle$,
639: $|t_l^0 \rangle = ( |\uparrow \downarrow \rangle
640:                    + |\downarrow \uparrow \rangle ) / \sqrt{2}$,
641: and $|t_l^- \rangle = |\downarrow \downarrow \rangle$.
642: When $h$ is small, the ground state consists of a product of
643: the singlet rungs.
644: As the field $h$ increases, the energy of the state
645: $|t_l^+ \rangle$ becomes lower,
646: and at $h = J_\perp(1+\Delta)/2$, the state degenerates with
647: $|s_l\rangle$.
648: We can thus analyze the low-energy properties of the system
649: for $h \simeq J_\perp(1+\Delta)/2$ by retaining only
650: the two lowest-energy states $|s_l\rangle$ and $|t_l^+ \rangle$
651: for each rung.
652: We may regard the two states as ^^ ^^ down" and ^^ ^^ up" states
653: of an effective $S = 1/2$ spin,
654: \begin{eqnarray}
655: \widetilde{S}^x_l&=&-\frac{1}{\sqrt{2}} (S^x_{1,l}-S^x_{2,l}),
656: \quad
657: \widetilde{S}^y_l=-\frac{1}{\sqrt{2}} (S^y_{1,l}-S^y_{2,l}),\\
658: \widetilde{S}^z_l&=&S^z_{1,l}+S^z_{2,l}-\frac{1}{2}.
659: \label{tilde S^z}
660: \end{eqnarray}
661: The effective spins $\widetilde{\bbox{S}}_l$ are the only low-energy
662: degrees of freedom, and their dynamics is governed by
663: \begin{eqnarray}
664: \widetilde{{\cal H}}
665:         &=&
666:    J_\parallel \sum_{l=1}^{L-1}
667:      (\widetilde{\bbox{S}}_l, \widetilde{\bbox{S}}_{l+1} )_{\Delta/2}
668:         - \frac{\Delta}{4} J_\parallel
669:                \left( \widetilde{S}_1^z + \widetilde{S}_L^z \right)
670:    \nonumber\\ &&
671:    - \left( h - \frac{1+\Delta}{2} J_\perp
672:             - \frac{\Delta}{2} J_\parallel
673:               \right) \sum_{l=1}^L \widetilde{S}_l^z
674:     + {\rm const}.
675:         \label{eq:Hefstr}
676: \end{eqnarray}
677: This mapping is derived in lowest order in $J_\parallel/J_\perp$ and
678: valid for the entire IC region of $0<M<1$, where $M$ is magnetization
679: per rung
680: \begin{equation}
681: M=\frac{1}{L}\sum_{l=1}^L\left(S^z_{1,l}+S^z_{2,l}\right).
682: \label{M}
683: \end{equation}
684: Note that the anisotropy of the effective $S=1/2$ $XXZ$ chain
685: is a half of the anisotropy of the original ladder, $\Delta/2$.
686: Besides the bulk effective magnetic field
687: $h - (1+\Delta) J_\perp /2 - \Delta J_\parallel /2$,
688: there is an additional field $- \Delta J_\parallel/4$
689: applied only to the boundary spins
690: $\widetilde{S}_1^z$ and $\widetilde{S}_L^z$.
691: It induces oscillating magnetization near the boundaries
692: superposed on the Friedel oscillation which is already present at
693: any $M\ne0$ without the boundary field.
694: The effect of the boundary field may be cancelled by adding an extra
695: term
696: \begin{equation}
697: {\cal H}' =
698:    h' \sum_{\mu = 1,2} \left( S^z_{\mu,1} + S^z_{\mu,L} \right)
699: \label{eq:H'}
700: \end{equation}
701: to the original ladder Hamiltonian (\ref{eq:Hlad}),
702: where $h' = \Delta J_\parallel/4$ for $j \gg 1$.
703: Now we define
704: \begin{eqnarray}
705: \bbox{S}_{0,l} &=& \bbox{S}_{1,l} + \bbox{S}_{2,l},
706:    \label{eq:S0} \\
707: \bbox{S}_{\pi,l} &=& \bbox{S}_{1,l} - \bbox{S}_{2,l}.
708:    \label{eq:Spi}
709: \end{eqnarray}
710:   From the mapping explained above, we conclude that the two-spin
711: correlations $\langle S_{\pi,l}^x S_{\pi,l'}^x \rangle$
712: and $\langle S_{0,l}^z S_{0,l'}^z \rangle$
713: and the local magnetization $\langle S_{0,l}^z\rangle$ in the open
714: ladder ${\cal H} + {\cal H}'$ in the limit $j\gg1$ are given by the
715: corresponding correlators in the $XXZ$ chain.
716: We thus obtain
717: \begin{eqnarray}
718: \langle S_{\pi,l}^x S_{\pi,l'}^x \rangle &=&
719:    2 X(l,l';Q),
720:         \label{eq:Cxlad} \\
721: \langle S_{0,l}^z S_{0,l'}^z \rangle &=&
722:    \left\langle
723:     \left(\frac{1}{2}+\widetilde{\bbox{S}}_l^z\right)
724:     \left(\frac{1}{2}+\widetilde{\bbox{S}}_{l'}^z\right)
725:    \right\rangle
726:    \nonumber\\
727:    &=& \frac{1}{4}+\frac{1}{2}[z(l;Q)+z(l';Q)]+Z(l,l';Q),
728:         \label{eq:Czlad} \\
729: \langle S_{0,l}^z\rangle &=&
730:    \frac{1}{2}+z(l,Q),  \label{eq:Szlad}
731: \end{eqnarray}
732: where the wavenumber is
733: \begin{equation}
734: Q = \frac{2\pi L}{L+1}\left(M-\frac{1}{2}\right).  \label{eq:Qlad}
735: \end{equation}
736: In the limit $L\to\infty$ the two-spin correlation functions reduce
737: to
738: \begin{eqnarray}
739: \langle S_{\pi,l}^x S_{\pi,l'}^x\rangle &=&
740: 2A_x\frac{(-1)^{l-l'}}{|l-l'|^{1/2K}} \nonumber\\
741: &&
742: -2\widehat{A}_x(-1)^{l-l'}
743:  \frac{\cos[2\pi M(l-l')]}{|l-l'|^{2K+(1/2K)}},
744: \\
745: \langle S_{0,l}^z S_{0,l'}^z\rangle &=&
746: M^2-\frac{1}{4\pi^2\eta|l-l'|^2} \nonumber\\
747: &&
748: +A_z\frac{\cos[2\pi M(l-l')]}{|l-l'|^{2K}}.
749: \end{eqnarray}
750: Note that they can be obtained from Eqs.~(\ref{eq:CxLchn}) and
751: (\ref{eq:CzLchn}) by replacing $q$ and $M_{\rm ch}$ with
752: $2\pi(M-1/2)$ and $M$, respectively.
753: On the other hand, the correlations
754: $\langle S_{0,l}^x S_{0,l'}^x \rangle$
755: and $\langle S_{\pi,l}^z S_{\pi,l'}^z \rangle$ decay exponentially
756: because $S^x_{0,l}$ and $S^z_{\pi,l}$ always create
757: the high-energy rung states
758: $|t_l^0 \rangle$ and $|t_l^- \rangle$
759: as a virtual excited state.
760: 
761: 
762: 
763: 
764: Next, we consider the opposite case, the weak-coupling limit ($j\ll1$).
765: The system in this limit has been investigated with the Abelian
766: bosonization method.\cite{Chi-Gia,Gia-Tsv,Furu-Zhn}
767: In these studies, two chains are first bosonized independently,
768: and then the interchain coupling $J_\perp$ is treated
769: perturbatively.\cite{Shel}
770: Four bosonic fields $\phi_\pm(x)$ and $\tilde{\phi}_\pm(x)$ are
771: introduced, where $\phi_+$ and $\tilde{\phi}_+$
772: ($\phi_-$ and $\tilde{\phi}_-$) are the symmetric (antisymmetric)
773: combinations of bosonic fields of each chain.
774: All the fields are massive\cite{Shel} when $h < h_{c1}$.
775: In the IC regime of $h_{c1} \le h \le h_{c2}$, on the other hand,
776: the fields $\phi_+$ and $\tilde{{\phi}}_+$ become massless
777: while the fields $\phi_-$ and $\tilde{{\phi}}_-$ remain massive.
778: The low-energy effective Hamiltonian for the gapless modes
779: has the same form
780: as that of the $S=1/2$ $XXZ$ chain, Eq.~(\ref{eq:HchnBos}).
781: Furthermore, the spin correlation functions
782: $\langle S_{\pi,l}^x S_{\pi,l'}^x \rangle$ and
783: $\langle S_{0,l}^z S_{0,l'}^z \rangle$ have the same $r$ dependence
784: as in the strong-coupling limit
785: (but with different values of $K$, $a$, $b$, and
786: $c$).\cite{Furu-Zhn,note}
787: The correlators
788: $\langle S_{0,l}^x S_{0,l'}^x \rangle$ and
789: $\langle S_{\pi,l}^z S_{\pi,l'}^z \rangle$
790: decay exponentially,\cite{Furu-Zhn} because they involve the massive
791: fields.
792: This result also matches the strong-coupling limit.
793: Moreover, the incommensurate wavenumber for the short-ranged
794: correlators is $\tilde q=\pi M$, which is different from the IC
795: wavenumber for the quasi-long-ranged correlators $q=2\pi(M-1/2)$.
796: For example, it was found that\cite{Furu-Zhn}
797: \begin{equation}
798: \langle S_{\pi,l}^z S_{\pi,l'}^z\rangle = \widetilde{A}_z
799: (-1)^{l-l'}e^{-|l-l'|/\xi}\frac{\cos[\pi M(l-l')]}{|l-l'|^{1/2+1/4\eta}},
800: \label{SpiSpi}
801: \end{equation}
802: where $\xi$ is a correlation length for a massive mode and
803: $\widetilde{A}_z$ is a constant.
804: 
805: 
806: 
807: 
808: We have seen that, both in the strong- and weak-coupling limits, the
809: low-energy physics of the two-leg ladder in the gapless regime is in
810: the same universality class as the $XXZ$ chain in a magnetic field.
811: In particular, the spin correlation functions
812: $\langle S^x_{\pi,l}S^x_{\pi,l'}\rangle$ and
813: $\langle S^z_{0,l}S^z_{0,l'}\rangle$ and the local magnetization
814: $\langle S^z_{0,l}\rangle$ have the same forms as the
815: corresponding functions in the $XXZ$ chain, but with the shifted
816: wavenumber $2\pi(M-1/2)$ and with different values of
817: $K$, $a$, $b$, and $c$.\cite{nonuniversal}
818: It is then very natural to postulate that the universality is not
819: restricted to the two limits but holds for any $j$.
820: This allows us to use Eqs.~(\ref{eq:Cxlad})--(\ref{eq:Szlad}) for
821: analyzing the correlation functions in the ladder for any $j$ and
822: $M$ ($0<M<1$).
823: We can thus determine the TL-liquid parameter $K$ of the ladder
824: by fitting the numerical data to
825: Eqs.~(\ref{eq:Cxlad})--(\ref{eq:Szlad}) in the same way as we did for
826: the $XXZ$ chain.
827: The result is presented in the next subsection.
828: Finally, we may expect that the $|l-l'|$ dependence of the short-ranged
829: correlators
830: $\langle S_{0,l}^x S_{0,l'}^x\rangle$ and
831: $\langle S_{\pi,l}^z S_{\pi,l'}^z\rangle$ obtained in the weak-coupling
832: analysis, such as Eq.\ (\ref{SpiSpi}), should also be valid for any $j$
833: and $M$ ($0<M<1$).
834: 
835: 
836: 
837: 
838: 
839: \subsection{Numerical Results}
840: Here we present the result of the DMRG calculation of the two-spin
841: correlations
842: $\langle S_{\pi,l}^x S_{\pi,l'}^x \rangle$ and
843: $\langle S_{0,l}^z S_{0,l'}^z \rangle$ and the local magnetization
844: $\langle S_{0,l}^z\rangle$ in the ladder.
845: Taking $R$, $a$, $b$, and $c$ as free parameters, we fit the data to
846: the formulas (\ref{eq:Cxlad})--(\ref{eq:Szlad}) for several values of
847: $j$ and estimate the $M$ dependence of $K = 1/(4 \pi R^2)$.
848: The numerical calculations were performed for the open ladder
849: ${\cal H} + {\cal H}'$ of $L=100$ rungs ($200$ sites)
850: for $j = 10.0, 2.0, 1.0, 0.5$ and $\Delta = 1.0, 0.5, 0.0$
851: using the finite-system DMRG method of the improved version.
852: We calculated the two-spin correlations for $l = r_0 - r/2$ and
853: $l' = r_0 + r/2$, where $r_0 = L/2$ for even $r$ and $r_0 = (L+1)/2$
854: for odd $r$.
855: The maximum value of the kept states $m$ is $160$.
856:   From the difference between the data with $m = 160$ and
857: those with $m = 120$, we estimate the numerical error
858: due to the truncation.
859: The estimated errors for $\langle S^x_{\pi,l} S^x_{\pi,l'} \rangle$,
860: $\langle S^z_{0,l} S^z_{0,l'} \rangle$,
861: and $\langle S^z_l \rangle$ are, at largest, of order
862: $10^{-4}$, $10^{-6}$, and $10^{-6}$,
863: which is almost negligible.
864: In the course of the calculation, we optimized the value of the extra
865: boundary field $h'$ to minimize
866: the effect of the boundary field.\cite{addH}
867: For finite $j$, however, the boundary effect cannot be eliminated
868: completely since it can be represented
869: by the form of ${\cal H}'$ only in the strong-coupling limit $j\gg1$.
870: As a result, the two-spin correlation functions and the local
871: magnetization in the open ladder ${\cal H} + {\cal H}'$ might deviate
872: from the expected form, Eqs.~(\ref{eq:Cxlad})--(\ref{eq:Szlad}),
873: near the boundaries.
874: For this reason we used data of smaller range of $r$ for the
875: fitting than in Sec.~II to reduce the unwanted boundary effect.
876: We chose the regions
877: $10 \le r \le 70$, $10 \le r \le 80$, $20 \le r \le 70$, and
878: $20 \le r \le 80$ for the fitting of the two-spin correlators
879: and $10 \le l \le 90$ and $20 \le l \le 80$ for the local magnetization.
880: As in Sec.~II, we regard the mean and
881: the variance of the fitting parameters
882: obtained for these different ranges of $r$ and $l$
883: as the estimated value and the error of the estimates,
884: respectively.
885: Incidentally, we have also checked for $(j,M)=(10.0, 0.5)$ that,
886: without the boundary field $h'$, the local magnetization
887: $\langle S^z_{0,l}\rangle$ has the Friedel oscillations induced by the
888: effective boundary field.
889: We found that the oscillations decay algebraically into the bulk with
890: the exponent $K$,
891: as expected from the bosonization analysis.\cite{Affleck}
892: 
893: 
894: 
895: 
896: The numerical data of $\langle S_{\pi,l}^x S_{\pi,l'}^x \rangle$,
897: $\langle S_{0,l}^z S_{0,l'}^z \rangle$, and $\langle S_{0,l}^z\rangle$
898: for $j = 10.0$ and $1.0$ with $\Delta = 1.0$ (Heisenberg case)
899: are shown in Figs.~\ref{fig:lad100} and \ref{fig:lad010}
900: by open symbols, whose sizes are larger than the truncation error
901: mentioned above.
902: The small solid symbols in the figures are the fits
903: to the DMRG data of the two-spin correlations for $10 \le r \le 80$
904: and to those of the local magnetization for $10 \le l \le 90$, respectively.
905: It is clearly seen that the fitting works extremely well
906: for $j = 10.0$, confirming the validity of the formulas
907: (\ref{eq:Cxlad}), (\ref{eq:Czlad}), and (\ref{eq:Szlad}).
908: Furthermore, the agreement between the numerical data and the fits
909: at $j = 1.0$ is also quite good except some deviations near the
910: boundary, indicating that the formulas are accurate for the
911: intermediate-coupling regime of $j$ as well.
912: We note that the quality of the fitting is also good for other values
913: of $j$ and $\Delta$ that we have examined.
914: We therefore conclude that the gapless mode of the two-leg ladders
915: is a TL liquid for arbitrary $j$,
916: and accordingly, the properties of the strong- and weak-coupling
917: ladders are smoothly connected.
918: 
919: 
920: 
921: 
922: Next we show in Fig.~\ref{fig:Klad} the $M$ dependence of
923: $K$ estimated from the data of
924: $\langle S_{\pi,l}^x S_{\pi,l'}^x \rangle$
925: for various $j$ in both the Heisenberg ($\Delta = 1.0$) and
926: $XY$ ($\Delta = 0$) cases.
927: In the earlier study\cite{Usami} the exponent $\eta$ was
928: obtained for $j=5.0$ in the Heisenberg case only.
929: We note that the estimation from
930: $\langle S_{\pi,l}^x S_{\pi,l'}^x \rangle$
931: is more reliable than that from
932: $\langle S_{0,l}^z S_{0,l'}^z \rangle$
933: or $\langle S_{0,l}^z\rangle$ as we have seen in the $XXZ$ chain.
934: Theoretically\cite{Furu-Zhn} it is expected that $K$ should
935: approach the universal value $K=1$ when $M\to0$ as well as when
936: $M\to1$, since the system is equivalent to the dilute limit of
937: hard-core bosons.
938: Although the data for $M\to0$ have large error bars,
939: we may conclude that our results for the Heisenberg case are
940: consistent with the theoretical prediction.
941: Our results for the $XY$ case show a more subtle feature.
942: At first sight the results for weaker couplings ($j=1.0$ and 0.5) do
943: not seem to approach $K=1$ as $M\to0$.
944: We think, however, that $K$ changes very rapidly at small $M$ to
945: approach $K=1$, in view of the data for $j=10.0$ and $2.0$, which
946: are consistent with the theory.
947: Unfortunately, it is difficult to numerically estimate $K$ for small
948: $M$ with high accuracy to resolve this issue.
949: 
950: 
951: 
952: In the strong-coupling limit $j\gg1$,
953: the ladder system with anisotropy $\Delta$ is
954: equivalent to the $S=1/2$ $XXZ$ chain with anisotropy $\Delta/2$,
955: as explained in the previous subsection.
956: Figure \ref{fig:Klad} clearly shows that
957: the estimated value of $K$ for $j = 10.0$ is consistent with
958: the anticipated behavior shown
959: as the dotted curves in both the Heisenberg and $XY$ cases.
960: As $j$ decreases, $K$ increases monotonically for any $M$ ($0<M<1$).
961: Thus, $K$ is always larger than 1 in the $XY$ ladder
962: because $K\to1$ for any $M$ in the large $j$ limit.
963: In the Heisenberg ladder, on the other hand, $K$ is smaller than 1 in
964: the strong-coupling limit, as expected from the mapping to the $XXZ$
965: chain.
966: Upon decreasing the interchain coupling $j$, $K$ starts to increase
967: and the $K$-$M$ relation changes from a concave curve to a convex one.
968: We note that the similar behavior is observed
969: also in the ladder with $\Delta = 0.5$: As $j$ decreases,
970: the $K$-$M$ relation changes from a concave curve at $j\gg1$,
971: corresponding to the behavior of the $XXZ$ chain
972: with anisotropy $\Delta/2 = 0.25$, to a convex one.
973: We thus consider that this behavior of the $K$-$M$ curve
974: is an universal feature for $0 < \Delta \le 1$.
975: 
976: 
977: 
978: The TL-liquid parameter $K$ determines the long-distance behavior of
979: correlation functions in the thermodynamic limit.
980: For example, the leading term of the correlator
981: $\langle S^z_{0,l} S^z_{0,l'}\rangle - M^2$
982: decays as $\cos[2\pi M(l-l')]/|l-l'|^{2K}$ for $K<1$,
983: while it decays like $|l-l'|^{-2}$ for $K>1$.
984: Hence, in the ladder with $0<\Delta\le1$
985: the leading term of the correlator changes
986: from $\cos[2\pi M(l-l')]/|l-l'|^{2K}$ to $|l-l'|^{-2}$
987: at a critical value $j_c(M)$ as $j$ decreases,
988: while in the $XY$ ladder the leading term
989: is always $|l-l'|^{-2}$.
990: On the other hand, the correlator
991: $\langle S^x_{\pi,l} S^x_{\pi,l'}\rangle$ decays as
992: $(-1)^{l-l'}/|l-l'|^{1/2K}$ in the whole range of $K$ covered
993: in Figs.~\ref{fig:Klad} (a) and \ref{fig:Klad} (b).
994: The temperature dependence of the spin-lattice relaxation rate $1/T_1$
995: in NMR experiments is directly related to the TL-liquid parameter $K$
996: through the decay exponent of the most slowly decaying
997: correlation.\cite{Chi-Gia}
998:   From the behavior of $K$-$M$ relation obtained above,
999: we find that the correlator $\langle S^x_{\pi,l} S^x_{\pi,l'}\rangle$
1000: decays most slowly for any $j$, $M$, and $0 \le \Delta \le 1$.
1001: We therefore conclude that at low temperatures the relaxation rate
1002: of the ladder in the gapless regime always shows a power-law divergence
1003: $1/T_1\propto T^{-1+(1/2K)}$.
1004: 
1005: 
1006: 
1007: 
1008: Figure \ref{fig:SzSz} shows the numerical result of
1009: $\langle S^z_{\pi,l}S^z_{\pi,l'}\rangle$ for the Heisenberg ladder
1010: at $j=0.5$.
1011: It exhibits exponentially decaying oscillatory behavior.
1012:  From the period $\lambda$ of oscillations,
1013: we obtain the IC wavenumber $\tilde q = 2\pi/\lambda$
1014: as a function of $M$; see the inset figure.
1015: The result confirms the theoretical prediction
1016: $\tilde q=\pi M$.
1017: This IC wavenumber $\tilde q$ tells us that the massive magnon
1018: dispersion has a minimum excitation energy at\cite{Furu-Zhn}
1019: $q=\pi-\tilde q=\pi(1-M)$.
1020: Accordingly, the dynamical spin structure factor
1021: $S^{zz}_\pi(q,\omega)$ should have a power-law divergence along the
1022: energy dispersion which is roughly shifted by $\pi M$ from that of
1023: the triplet magnon dispersion in the absence of the magnetic field.
1024: It would be interesting if this feature is observed by inelastic
1025: neutron scattering experiments.
1026: 
1027: 
1028: 
1029: 
1030: 
1031: 
1032: \section{CONCLUSIONS}
1033: In this paper we have studied the ground-state spin correlations
1034: in the gapless IC regime of the $S=1/2$ $XXZ$ chain and the two-leg
1035: AF ladder in a magnetic field.
1036: We have used the $S=1/2$ $XXZ$ chain as a first test ground to apply
1037: the method we developed in our previous work:
1038: We numerically computed the two-spin correlation functions and the
1039: local magnetization by the DMRG method and fit the results to
1040: functions which are obtained using the bosonization technique.
1041: The fitting parameters are the TL-liquid parameter $K$ and the
1042: amplitudes of bosonic operators.
1043: We found good agreement between $K$ estimated from the fitting and
1044: $K$ calculated from the Bethe ansatz.
1045: As a byproduct we obtained the amplitudes of the dominant terms in
1046: $\langle S^x_l S^x_{l'}\rangle$ and $\langle S^z_l S^z_{l'}\rangle$.
1047: 
1048: 
1049: 
1050: 
1051: We have applied the same technique to
1052: the two-leg AF ladder in the gapless IC regime.
1053: It has been known that in both the strong- and weak-coupling limits
1054: the low-energy excitations in the ladder are regarded
1055: as a TL liquid like the $XXZ$ chain in a field.
1056: We fit our DMRG data of the two-spin correlation functions and the
1057: local magnetization of the ladder to the same bosonization
1058: formulas we used in the analysis of the $XXZ$ chain.
1059: The fitting worked very well not only in the strong- and weak-coupling
1060: limits but for broad range of the interchain coupling strength $j$.
1061: We thereby confirmed that the low-energy gapless excitations are
1062: indeed described as the TL liquid for any $j$ and
1063: the properties of the strong- and weak-coupling ladders
1064: are smoothly connected.
1065: For several values of $j$, we have determined $K$,
1066: which shows nontrivial $j$ and $M$ dependences
1067: (Fig.~\ref{fig:Klad}).
1068: It turned out that, for any $M$ ($0<M<1$),
1069: $K$ is a monotonically decreasing function of $j$.
1070: In the ladder with anisotropy $0 < \Delta \le 1$
1071: the $K$-$M$ relation changes from a concave curve at $j\gg1$
1072: to a convex one as $j$ decreases,
1073: while in the $XY$ ladder ($\Delta=0$)
1074: it changes from a line $K=1$ at $j\gg1$ to a convex curve.
1075: We also found that the spin-lattice relaxation rate
1076: in NMR measurement shows a power-law divergence
1077: $1/T_1 \propto T^{-1+(1/2K)}$ at low temperature for any $j$.
1078: 
1079: 
1080: 
1081: 
1082: 
1083: 
1084: \acknowledgements
1085: 
1086: 
1087: 
1088: 
1089: Numerical computations were performed at the Yukawa Institute
1090: Computing Facility.
1091: The work of AF was in part supported by Grant-in-Aid for Scientific
1092: Research on Priority Areas (A) from the Ministry of Education,
1093: Science, Sports and Culture (No.~12046238) and by Grant-in-Aid for
1094: Scientific Research from the Japan Society for the Promotion of
1095: Science (No.~11740199).
1096: 
1097: 
1098: 
1099: 
1100: \appendix
1101: \section*{Derivation of Correlators}
1102: 
1103: 
1104: 
1105: 
1106: We briefly explain the derivation of Eqs.~(\ref{eq:Cxechn}) and
1107: (\ref{eq:Czechn}).
1108: As is always the case  with the bosonization, we need to introduce a
1109: short-distance cutoff to obtain finite results.
1110: The lattice spacing in the original Hamiltonian serves as
1111: the natural cutoff scale.
1112: 
1113: 
1114: 
1115: 
1116: When we use Eqs.~(\ref{eq:Szchn}) and (\ref{eq:S-chn}) with the mode
1117: expansions (\ref{eq:mode1}) and (\ref{eq:mode2}), we encounter the
1118: summation
1119: \begin{equation}
1120: \sum^\infty_{n=1}\frac{1}{n}
1121:    \left[1-\cos\left(\frac{\pi nl}{L+1}\right)\right],
1122: \end{equation}
1123: which is formally divergent.
1124: We regularize it by inserting an exponential factor
1125: $e^{-\pi n/(L+1)}$:
1126: \begin{equation}
1127: \sum^\infty_{n=1}\frac{1}{n}e^{-\pi n/(L+1)}
1128: \left[1-\cos\left(\frac{\pi nl}{L+1}\right)\right]
1129: =\ln[f(l)],
1130: \label{ln(f)}
1131: \end{equation}
1132: where $f(l)\equiv f_1(l)$ defined in Eq.~(\ref{eq:fx}).
1133: We note that Eq.~(\ref{ln(f)}) is a very good approximation except
1134: near the points where $f_1(l)$ is divergent.
1135: Taking derivatives with respect to $l$, we obtain
1136: \begin{equation}
1137: \sum^\infty_{n=1}e^{-\pi n/(L+1)}\sin\left(\frac{\pi nl}{L+1}\right)
1138: =\frac{1}{2}\cot\left(\frac{\pi l}{2(L+1)}\right)
1139: \end{equation}
1140: and
1141: \begin{equation}
1142: \sum^\infty_{n=1}n e^{-\pi n/(L+1)}\cos\left(\frac{\pi nl}{L+1}\right)
1143: =-\frac{1}{4\sin^2\left(\frac{\pi l}{2(L+1)}\right)}.
1144: \end{equation}
1145: Another point to note is that for $\varepsilon_i=\pm1$
1146: \begin{eqnarray}
1147: \langle e^{i2\pi R\epsilon_1\tilde\phi(l)}
1148:    e^{i2\pi R\epsilon_2\tilde\phi(l')}\rangle
1149: &\propto&
1150:    \int^{1/R}_0 R e^{i2\pi R(\epsilon_1+\epsilon_2)\tilde\phi_0}
1151:     d\tilde\phi_0 \nonumber\\
1152: &\propto&
1153: \delta_{\epsilon_1+\epsilon_2,0}.
1154: \end{eqnarray}
1155: 
1156: 
1157: 
1158: 
1159: With the above-mentioned formulas, it is straightforward to obtain
1160: \begin{eqnarray*}
1161: &&
1162: \langle\cos[2\pi R\tilde\phi(l)]\cos[2\pi R\tilde\phi(l')]\rangle
1163: =\frac{[f(2l)f(2l')]^{\eta/2}}{2[f(l-l')f(l+l')]^\eta},
1164:    \\
1165: &&
1166: \langle e^{i2\pi R\epsilon_1\tilde\phi(l)}
1167:    e^{-i2\pi R\epsilon_1\tilde\phi(l')}e^{i\epsilon_2\phi(l')/R}\rangle
1168: \\
1169: &&
1170: =i\epsilon_1\epsilon_2 e^{iq\epsilon_2 l'}
1171: \frac{{\rm sgn}(l-l')[f(2l)f(2l')]^{\eta/2}}
1172:        {[f(l-l')f(l+l')]^\eta [f(2l')]^{1/2\eta}},
1173: \\
1174: &&
1175: \langle e^{i2\pi R\epsilon_0\tilde\phi(l)}e^{i\epsilon_1\phi(l)/R}
1176:    e^{-i2\pi R\epsilon_0\tilde\phi(l')}e^{i\epsilon_2\phi(l')/R}\rangle
1177: \\
1178: &&
1179: =-\epsilon_1\epsilon_2 e^{iq(\epsilon_1l+\epsilon_2l')}
1180: \frac{[f(2l)f(2l')]^{\eta/2-1/2\eta}}
1181:        {[f(l-l')f(l+l')]^\eta}
1182: \left(\frac{f(l-l')}{f(l+l')}\right)^{\epsilon_1\epsilon_2/\eta},
1183: \\
1184: &&
1185: \left\langle\frac{d\phi}{dl}\frac{d\phi}{dl'}\right\rangle
1186: =-\frac{1}{2\pi}\left(\frac{1}{f_2(l-l')}+\frac{1}{f_2(l+l')}\right)
1187: +\left(\frac{q}{2\pi}\right)^2,
1188: \\
1189: &&
1190: \left\langle\left(\frac{d\phi}{dl}-\frac{q}{2\pi}\right)
1191:               \sin\frac{\phi(l')}{R}\right\rangle
1192: =\frac{\cos(ql')}{2\pi R}\frac{g(l+l')-g(l-l')}{[f(2l')]^{1/2\eta}},
1193: \\
1194: &&
1195: \langle e^{i\epsilon_1\phi(l)/R} e^{i\epsilon_2\phi(l')/R}\rangle
1196: =\frac{e^{iq(\epsilon_1l+\epsilon_2l')}}{[f(2l)f(2l')]^{1/2\eta}}
1197:    \left(\frac{f(l-l')}{f(l+l')}\right)^{\epsilon_1\epsilon_2/\eta}.
1198:    \\
1199: \end{eqnarray*}
1200: 
1201: 
1202: 
1203: 
1204: 
1205: 
1206: \begin{references}
1207: \bibitem{review}
1208: For a review, see, {\it e.g.}, E. Dagotto and T.M. Rice,
1209:    Science {\bf 271}, 618 (1996).
1210: 
1211: 
1212: 
1213: 
1214: \bibitem{RVB}
1215: P. W. Anderson, Science {\bf 235}, 1196 (1987).
1216: 
1217: 
1218: 
1219: 
1220: \bibitem{exp1}
1221: M. Azuma, Z. Hiroi, M. Takano, K. Ishida, and Y. Kitaoka,
1222:    Phys. Rev. Lett. {\bf 73}, 3463 (1994).
1223: 
1224: 
1225: 
1226: 
1227: \bibitem{exp2}
1228: G. Chaboussant, P.A. Crowell, L.P. L\'evy, O. Piovesana, A. Madouri,
1229: and D. Mailly, Phys. Rev. B {\bf 55}, 3046 (1997);
1230: G. Chaboussant, Y. Fagot-Revurat, M.-H. Julien, M.E. Hanson,
1231: C. Berthier, M. Horvati\'c, L.P. L\'evy, and O. Piovesana,
1232: Phys. Rev. Lett. {\bf 80}, 2713 (1998).
1233: 
1234: 
1235: 
1236: 
1237: \bibitem{Chi-Gia}
1238: R. Chitra and T. Giamarchi, Phys. Rev. B {\bf 55}, 5816 (1997).
1239: They concluded incorrectly that the incommensurate correlations in a
1240: spin ladder decay exponentially and the power-law components appear
1241: only at $q=0$ and $\pi$.
1242: This was corrected in Refs.~\onlinecite{Furu-Zhn} and
1243: \onlinecite{Gia-Tsv}.
1244: 
1245: 
1246: 
1247: 
1248: \bibitem{Gia-Tsv}
1249: T. Giamarchi and A.M. Tsvelik, Phys. Rev. B {\bf 59}, 11398 (1999).
1250: 
1251: 
1252: 
1253: 
1254: \bibitem{Furu-Zhn}
1255: A. Furusaki and S.C. Zhang, Phys. Rev. B {\bf 60}, 1175 (1999).
1256: 
1257: 
1258: 
1259: 
1260: \bibitem{CorAm}
1261: T. Hikihara and A. Furusaki, Phys. Rev. B {\bf 58}, R583 (1998).
1262: 
1263: 
1264: 
1265: 
1266: \bibitem{White1}
1267: S.R. White, Phys. Rev. Lett. {\bf 69}, 2863 (1992).
1268: 
1269: 
1270: 
1271: 
1272: \bibitem{White2}
1273: S.R. White, Phys. Rev. B {\bf 48}, 10345 (1993).
1274: 
1275: 
1276: 
1277: 
1278: \bibitem{Bethe0}
1279: C.N. Yang and C.P. Yang, Phys. Rev. {\bf 150}, 321 (1966).
1280: 
1281: 
1282: 
1283: 
1284: \bibitem{Bethe1}
1285: F.D.M. Haldane, Phys. Rev. Lett. {\bf 45}, 1358 (1980).
1286: 
1287: 
1288: 
1289: 
1290: \bibitem{Bethe2}
1291: F. Woynarovich, H.-P. Eckle, and T.T. Truong,
1292:    J. Phys. A: Math. Gen. {\bf 22}, 4027 (1989)
1293: 
1294: 
1295: 
1296: 
1297: \bibitem{Eggert}
1298: S. Eggert and I. Affleck, Phys. Rev. B {\bf 46}, 10866 (1992).
1299: 
1300: 
1301: 
1302: 
1303: \bibitem{Friedel1}
1304: M. Fabrizio and A.O. Gogolin, Phys. Rev. B {\bf 51}, 17827 (1995).
1305: 
1306: 
1307: 
1308: 
1309: \bibitem{Friedel2}
1310: R. Egger and H. Grabert, Phys. Rev. Lett. {\bf 75}, 3505 (1995).
1311: 
1312: 
1313: 
1314: 
1315: \bibitem{Friedel3}
1316: S. Eggert, H. Johannesson, and A. Mattsson,
1317:    Phys. Rev. Lett. {\bf 76}, 1505 (1996).
1318: 
1319: 
1320: 
1321: 
1322: \bibitem{Friedel4}
1323: G. Bed\"urftig, B. Brendel, H. Frahm,  and R.M. Noack,
1324:    Phys. Rev. B {\bf 58}, 10225 (1998).
1325: 
1326: 
1327: 
1328: 
1329: \bibitem{TLpr1}
1330: N.M. Bogoliubov, A.G. Izergin, and V.E. Korepin,
1331:                Nucl. Phys. B {\bf 275}, 687 (1986).
1332: 
1333: 
1334: 
1335: 
1336: \bibitem{TLpr2}
1337: D.C. Cabra, A. Honecker, and P. Pujol,
1338:    Phys. Rev. B {\bf 58}, 6241 (1998).
1339: 
1340: 
1341: 
1342: 
1343: \bibitem{White3}
1344: S.R. White, Phys. Rev. Lett. {\bf 77}, 3633 (1996).
1345: 
1346: 
1347: 
1348: 
1349: \bibitem{Lu-Za}
1350: S. Lukyanov and A. Zamolodchikov, Nucl. Phys. B {\bf 493}, 571 (1997).
1351: 
1352: 
1353: 
1354: 
1355: \bibitem{Luky}
1356: S. Lukyanov, Phys. Rev. B {\bf 59}, 11163 (1999).
1357: 
1358: 
1359: 
1360: 
1361: \bibitem{no hat Ax}
1362: We are not sure about the accuracy of our estimates for the amplitude
1363: $\widehat{A}_x$ of the subleading term, and thereby we do not show
1364: them here.
1365: 
1366: 
1367: 
1368: 
1369: \bibitem{Ampnote}
1370: The estimates of $A_z$ for small $M>0$ shown in Table \ref{tab:Achn} (b)
1371: do not seem to approach smoothly the exact value at $M_{\rm ch}=0$
1372: as $M_{\rm ch} \to 0$.
1373: We do not know exactly the reason why this happens.
1374: This might be due to  the same reason as the one for the deviation
1375: of $K$ estimated from $\langle S^z_l \rangle$ mentioned in the text.
1376: 
1377: 
1378: 
1379: \bibitem{Mila}
1380: F. Mila, Eur. Phys. J. B {\bf 6}, 201 (1998).
1381: 
1382: 
1383: 
1384: 
1385: \bibitem{Totsu}
1386: K. Totsuka, Phys. Rev. B {\bf 57}, 3454 (1998).
1387: 
1388: 
1389: 
1390: 
1391: \bibitem{Shel}
1392: D.G. Shelton, A.A. Nersesyan, and A.M. Tsvelik,
1393: Phys. Rev. B {\bf 53}, 8521 (1996).
1394: 
1395: 
1396: 
1397: 
1398: \bibitem{note}
1399: We note that the notation has changed from Ref.\ \onlinecite{Furu-Zhn}.
1400: The parameter $\eta$ in the present paper is equal to the inverse of
1401: the $\eta$ used in Ref.\ \onlinecite{Furu-Zhn}.
1402: 
1403: 
1404: 
1405: 
1406: \bibitem{nonuniversal}
1407: In the strong-coupling limit, the parameters $K$, $a$, $b$, and $c$
1408: in the ladder with the anisotropy $\Delta$ are
1409: related to those in the $XXZ$ chain with the anisotropy $\Delta/2$;
1410: see Eqs.~(\ref{eq:Cxlad})-(\ref{eq:Qlad}).
1411: 
1412: 
1413: 
1414: \bibitem{addH}
1415: To be more concrete, we optimized the value of $h'$ to
1416: minimize the deviation of
1417: $\langle S^z_{0,l} \rangle$ for $M = 1/2$ from the constant value,
1418: $1/2$; see Eq.~(\ref{eq:Szlad}).
1419: 
1420: 
1421: 
1422: 
1423: \bibitem{Affleck}
1424: I. Affleck, J. Phys. A: Math. Gen. {\bf 31}, 2761 (1998).
1425: 
1426: 
1427: 
1428: 
1429: \bibitem{Usami} M. Usami and S. Suga,
1430: Phys. Rev. B {\bf 58}, 14401 (1998).
1431: 
1432: 
1433: 
1434: 
1435: \end{references}
1436: 
1437: 
1438: 
1439: \end{multicols}
1440: 
1441: 
1442: 
1443: \begin{table}
1444: \caption{
1445: The correlation amplitudes;
1446: (a) $A_x = c^2/2$ estimated from the data of
1447: $\langle S^x_l S^x_{l'} \rangle$;
1448: (b) $A_z = a^2/2$ estimated from the data of
1449: $\langle S^z_l \rangle$.
1450: The figures in parentheses indicate the error bar
1451: on the last quoted digits.
1452: The error bars of $A_z$ for $\Delta=0$ and $0.05 \le M \le 0.45$
1453: are smaller than $10^{-5}$.
1454: The exact values for $M_{\rm ch} = 0$ given by
1455: Eqs. (\ref{eq:LZx}) and (\ref{eq:LZz}) are also listed.
1456: }
1457: \label{tab:Achn}
1458: (a)$A_x$
1459: \begin{tabular}{ccccccccccc}
1460: $M_{\rm ch}$    &    0      &    0.05   &    0.10   &    0.15   &    0.20
1461:    &    0.25   &    0.30   &    0.35   &    0.40   &    0.45   \\ \hline
1462: $\Delta= 0.0$ & 0.14709   & 0.14626(1)& 0.14364(6)& 0.1390(1) & 0.13262(1)
1463:    & 0.12410(6)& 0.1132(3) & 0.0993(7) & 0.081(2)  & 0.0594(7) \\
1464: $\Delta= 0.1$ & 0.14451   & 0.14369(7)& 0.1413(1) & 0.1371(1) & 0.13101(7)
1465:    & 0.12293(2)& 0.1125(3) & 0.0991(7) & 0.081(2)  & 0.0597(7) \\
1466: $\Delta= 0.2$ & 0.14187   & 0.1408(4) & 0.1390(2) & 0.1351(2) & 0.1294(1)
1467:    & 0.12174(5)& 0.1111(7) & 0.0988(6) & 0.081(2)  & 0.0600(7) \\
1468: $\Delta= 0.3$ & 0.13921   & 0.1384(3) & 0.1366(3) & 0.1330(3) & 0.1278(2)
1469:    & 0.12053(9)& 0.1111(2) & 0.0985(6) & 0.081(2)  & 0.0601(7) \\
1470: $\Delta= 0.4$ & 0.13656   & 0.1358(3) & 0.1342(4) & 0.1310(3) & 0.1261(3)
1471:    & 0.1193(1) & 0.1104(1) & 0.0982(6) & 0.081(2)  & 0.0603(7) \\
1472: $\Delta= 0.5$ & 0.13400   & 0.1332(4) & 0.1318(5) & 0.1289(4) & 0.1245(3)
1473:    & 0.1182(2) & 0.10973(9)& 0.0979(5) & 0.081(2)  & 0.0605(7) \\
1474: $\Delta= 0.6$ & 0.13164   & 0.1310(5) & 0.1294(6) & 0.1268(5) & 0.1229(4)
1475:    & 0.1170(2) & 0.10905(7)& 0.0976(5) & 0.081(2)  & 0.0606(7) \\
1476: $\Delta= 0.7$ & 0.12973   & 0.1281(6) & 0.1270(7) & 0.1248(5) & 0.1213(4)
1477:    & 0.1159(3) & 0.10839(6)& 0.0973(5) & 0.081(2)  & 0.0607(7) \\
1478: $\Delta= 0.8$ & 0.12896   & 0.1257(8) & 0.1247(8) & 0.1227(6) & 0.1197(5)
1479:    & 0.1148(3) & 0.10775(7)& 0.0970(5) & 0.081(1)  & 0.0609(7) \\
1480: $\Delta= 0.9$ & 0.13214   & 0.1233(9) & 0.1223(9) & 0.1207(7) & 0.1182(6)
1481:    & 0.1137(3) & 0.10714(8)& 0.0967(4) & 0.081(1)  & 0.0610(8) \\
1482: $\Delta= 1.0$ &           & 0.121(1)  & 0.120(1)  & 0.1188(8) & 0.1177(9)
1483:    & 0.1127(4) & 0.1065(1) & 0.0958(6) & 0.081(1)  & 0.0610(8)
1484: \end{tabular}
1485: (b)$A_z$
1486: \begin{tabular}{ccccccccccc}
1487: $M_{\rm ch}$    &    0      &    0.05   &    0.10   &    0.15   &    0.20
1488:      &    0.25   &    0.30   &    0.35   &    0.40   &    0.45   \\ \hline
1489: $\Delta= 0.0$  & 0.05066   & 0.05066   & 0.05066   & 0.05066   & 0.05066
1490:      & 0.05066   & 0.05066   & 0.05066   & 0.05066   & 0.05066   \\
1491: $\Delta= 0.1$  & 0.05929   & 0.0599(7) & 0.0581(3) & 0.0567(6) & 0.0544(1)
1492:      & 0.0537(7) & 0.0513(6) & 0.0516(6) & 0.049(1)  & 0.0510(8) \\
1493: $\Delta= 0.2$  & 0.06891    & 0.071(1)  & 0.0662(6) & 0.063(1)  & 0.0580(2)
1494:      & 0.056(1)  & 0.052(1)  & 0.052(1)  & 0.048(2)  & 0.051(1)  \\
1495: $\Delta= 0.3$  & 0.07978    & 0.083(2)  & 0.0748(7) & 0.069(1)  & 0.0614(4)
1496:      & 0.059(1)  & 0.052(1)  & 0.053(1)  & 0.048(2)  & 0.052(2)  \\
1497: $\Delta= 0.4$  & 0.09231   & 0.097(3)  & 0.0838(6) & 0.075(1)  & 0.0645(6)
1498:      & 0.060(1)  & 0.053(2)  & 0.053(2)  & 0.047(3)  & 0.052(2)  \\
1499: $\Delta= 0.5$  & 0.10713   & 0.113(5)  & 0.093(4)  & 0.080(1)  & 0.0674(8)
1500:      & 0.062(2)  & 0.053(2)  & 0.053(2)  & 0.046(3)  & 0.052(3)  \\
1501: $\Delta= 0.6$  & 0.12539   & 0.132(6)  & 0.10263(5)& 0.0854(9) & 0.070(1)
1502:      & 0.063(2)  & 0.054(2)  & 0.053(2)  & 0.046(3)  & 0.052(3)  \\
1503: $\Delta= 0.7$  & 0.14930   & 0.153(8)  & 0.1121(5) & 0.0903(4) & 0.072(1)
1504:      & 0.065(2)  & 0.054(2)  & 0.053(2)  & 0.045(4)  & 0.052(3)  \\
1505: $\Delta= 0.8$  & 0.18414   & 0.176(10) & 0.121(1)  & 0.09486(6)& 0.074(2)
1506:      & 0.066(2)  & 0.054(2)  & 0.054(2)  & 0.045(4)  & 0.052(4)  \\
1507: $\Delta= 0.9$  & 0.24844   & 0.20(1)   & 0.131(2)  & 0.0990(7) & 0.076(2)
1508:      & 0.067(1)  & 0.054(2)  & 0.054(2)  & 0.047(4)  & 0.052(4)  \\
1509: $\Delta= 1.0$  &           & 0.23(1)   & 0.139(3)  & 0.103(1)  & 0.078(2)
1510:      & 0.067(1)  & 0.054(3)  & 0.054(2)  & 0.044(4)  & 0.052(4)
1511: \end{tabular}
1512: \end{table}
1513: 
1514: 
1515: 
1516: 
1517: \newpage
1518: \begin{multicols}{2}
1519: 
1520: 
1521: 
1522: \begin{figure}
1523: \epsfxsize=80mm
1524: \epsfbox{fig1a.eps}
1525: \epsfxsize=80mm
1526: \epsfbox{fig1b.eps}
1527: \epsfxsize=80mm
1528: \epsfbox{fig1c.eps}
1529: \narrowtext
1530: \caption{(a) $(-1)^{|l-l'|} \langle S_l^x S_{l'}^x \rangle$
1531: versus $r = |l-l'|$,
1532: (b) $|\langle S_l^z S_{l'}^z \rangle|$ versus $r$,
1533: (c) $\langle S_l^z \rangle$ versus $l$ for $\Delta = 0.5$
1534: and $M_{\rm ch}=0.05$, 0.25, and 0.45.
1535: The open symbols are the DMRG data and small solid symbols are
1536: the fits.
1537: The numerical errors of the DMRG data are smaller than the size of
1538: the open symbols.
1539: The data of $(-1)^{|l-l'|} \langle S_l^x S_{l'}^x \rangle$
1540: for $M_{\rm ch}=0.05$ and $0.45$ in figure (a)
1541: are multiplied by a factor of $2$ and $0.5$, respectively.}
1542: \label{fig:chn}
1543: \end{figure}
1544: 
1545: 
1546: 
1547: 
1548: \begin{figure}
1549: \epsfxsize=80mm
1550: \epsfbox{fig2.eps}
1551: \caption{The results of $K = 1/(4 \pi R^2)$ estimated from the fitting
1552: of $\langle S^x_l S^x_{l'} \rangle$ for $0 \le \Delta \le 1.0$.
1553: The dotted curves represent the exact values obtained from Bethe ansatz.}
1554: \label{fig:Kchn}
1555: \end{figure}
1556: 
1557: 
1558: 
1559: 
1560: \begin{figure}
1561: \epsfxsize=80mm
1562: \epsfbox{fig3a.eps}
1563: \epsfxsize=80mm
1564: \epsfbox{fig3b.eps}
1565: \epsfxsize=80mm
1566: \epsfbox{fig3c.eps}
1567: \caption{(a) $(-1)^{|l-l'|} \langle S_{\pi,l}^x S_{\pi,l'}^x \rangle$
1568: versus $r = |l-l'|$,
1569: (b) $|\langle S_{0,l}^z S_{0,l'}^z \rangle|$ versus $r$,
1570: (c) $\langle S_{0,l}^z \rangle$ versus $l$ for
1571: $j = 10.0$ and the Heisenberg case ($\Delta = 1.0$).
1572: The open symbols are the DMRG data and small solid symbols are
1573: the fits.
1574: The numerical errors of the DMRG data are smaller than the size of
1575: the open symbols.
1576: The data of $(-1)^{|l-l'|} \langle S_{\pi,l}^x S_{\pi,l'}^x \rangle$
1577: for $M=0.10$ and $0.90$ in figure (a)
1578: are multiplied by a factor of $2$ and $0.5$, respectively.}
1579: \label{fig:lad100}
1580: \end{figure}
1581: 
1582: 
1583: 
1584: 
1585: \begin{figure}
1586: \epsfxsize=80mm
1587: \epsfbox{fig4a.eps}
1588: \epsfxsize=80mm
1589: \epsfbox{fig4b.eps}
1590: \epsfxsize=80mm
1591: \epsfbox{fig4c.eps}
1592: \caption{(a) $(-1)^{|l-l'|} \langle S_{\pi,l}^x S_{\pi,l'}^x \rangle$
1593: versus $r = |l-l'|$,
1594: (b) $|\langle S_{0,l}^z S_{0,l'}^z \rangle|$ versus $r$,
1595: (c) $\langle S_{0,l}^z \rangle$ versus $l$ for
1596: $j = 1.0$ and the Heisenberg case ($\Delta = 1.0$).
1597: The open symbols are the DMRG data and small solid symbols are
1598: the fitting results.
1599: The numerical errors of the DMRG data are smaller than the size of
1600: the open symbols.
1601: The data of $(-1)^{|l-l'|} \langle S_{\pi,l}^x S_{\pi,l'}^x \rangle$
1602: for $M=0.10$ and $0.90$ in figure (a)
1603: are multiplied by a factor of $2$ and $0.5$, respectively.}
1604: \label{fig:lad010}
1605: \end{figure}
1606: 
1607: 
1608: 
1609: 
1610: \begin{figure}
1611: \epsfxsize=80mm
1612: \epsfbox{fig5a.eps}
1613: \epsfxsize=80mm
1614: \epsfbox{fig5b.eps}
1615: \caption{The TL-liquid parameter $K = 1/(4 \pi R^2)$ estimated from
1616: the fitting of $\langle S^x_{\pi,l} S^x_{\pi,l'} \rangle$ for
1617: $j = 10.0, 2.0, 1.0, 0.5$ and
1618: for (a) the Heisenberg ladder ($\Delta = 1.0$) and
1619: (b) the $XY$ ladder ($\Delta = 0$).
1620: The dotted curves represent the exact values of the $S=1/2$ $XXZ$
1621: chain with the anisotropy $\Delta/2$.}
1622: \label{fig:Klad}
1623: \end{figure}
1624: 
1625: 
1626: 
1627: 
1628: \begin{figure}
1629: \epsfxsize=80mm
1630: \epsfbox{fig6.eps}
1631: \caption{The correlation function
1632: $(-1)^{|l-l'|} \langle S^z_{\pi,l}S^z_{\pi,l'}\rangle$ 
1633: in the Heisenberg ladder at $j=0.5$ and $M=0.2$.
1634: The dotted curve is a guide to the eye.
1635: Inset: The $M$ dependence of the IC wavenumber $\tilde{q}$
1636: of $\langle S^z_{\pi,l}S^z_{\pi,l'}\rangle$ in the Heisenberg ladder
1637: with $j = 0.5$.
1638: The dotted line represents the theoretical prediction
1639: $\tilde{q}/\pi = M$.
1640: }
1641: \label{fig:SzSz}
1642: \end{figure}
1643: 
1644: 
1645: 
1646: \end{multicols}
1647: 
1648: 
1649: 
1650: \end{document}