您现在的位置:首页 > >

Ni-Impurity Effects on Incommensurate Spin Correlations in Superconducting La2-xSrxCuO4 (x=

发布时间:

Typeset with jpsj2.cls <ver.1.2>

Full Paper

Ni-Impurity E?ects on Incommensurate Spin Correlations in Superconducting La2?x Srx CuO4 (x = 0.06 and 0.07)

arXiv:cond-mat/0611123v1 [cond-mat.supr-con] 5 Nov 2006

Haruhiro Hiraka ? , Soichi Ohta1 , Shuich Wakimoto2 , Masaaki Matsuda2 and Kazuyoshi Yamada
Institute for Materials Research, Tohoku University, Sendai 980-8577 Department of Physics, Graduate School of Science, Tohoku University, Sendai 980-8578 2 Quantum Beam Science Directorate, Japan Atomic Energy Agency, Tokai, Ibaraki 319-1195
1

Neutron scattering experiments have been carried out to explore Ni-impurity e?ects on static spin correlations in La2?x Srx CuO4 (LSCO) in the vicinity of the superconductor-insulator boundary where both parallel and diagonal spin-density modulations (SDM) coexist at low temperature. Upon dilute Ni substitution the incommensurability decreases for both types of SDM, while the volume fraction of the diagonal (parallel) SDM increases (decreases). Subsequent Ni doping induces a bulk three-dimensional antiferromagnetic (AF) order when x ? Ni concentration. TN of such the AF order depends on x and seems to disappear at x ? 0.1. These e?ects are approximately ascribed by a reduction of mobile holes, and by a transition from the parallel to the diagonal SDM induced by Ni. KEYWORDS: La2?x Srx CuO4 , neutron scattering, spin density modulation, substitution e?ect, Ni, incommensurate spin correlations

1. Introduction Doped high-Tc cuprates, one of the most fruitful examples of doped Mott insulators, provide us rich information on the interplay between the doped carriers and the spin correlations commonly underlying on the Cu-O square lattices. Through a series of systematic studies, we have discovered clear relationships between the doping dependence of spin correlation and the onset of the high-Tc superconductivity in both underdoped1 and overdoped superconducting phases.2 In the superconducting (SC) phase, the so-called parallel spindensity modulations (P-SDM) commonly exist in holedoped LSCO. Therefore, the discovery of the so-called diagonal spin-density modulations (D-SDM) in the insulating spin-glass (SG) phase by Wakimoto et al.3 strongly suggested a transition of spin correlation between the DSDM and P-SDM at the underdoped SG-SC boundary. In fact, the detailed study in the vicinity of the boundary between SG and SC phases con?rmed a D-SDM to P-SDM transition at the boundary.4–6 To clarify the origin of the D-SDM we studied the impurity e?ect in the SG phase.7 The results show that Ni doping quickly destroys the incommensurability and restores the N?el ordering, indicating a strong e?ect on e hole localization. This suggests that Ni is doped as Ni3+ or as Ni2+ with a hole forming a strongly bound state. Therefore, Ni doping reduces the number of mobile or hopping Zhang-Rice (ZR) singlet states around Cu spins by creating localized hole sites near the doped Ni. Then the concentration of the mobile ZR singlet (xe? ) can be described by the di?erence between the number of holes and doped Ni ions. In fact, the xe? dependences of the incommensurability and the onset temperature of the DSDM for the Ni doped samples can be plot on the same phase diagram without impurities. This means that the
? E-mail

incommensurability in this system is controlled by the number of mobile ZR singlets or mobile holes. The similar localization e?ect around Ni impurities is also observable in lightly doped antiferromagnetic (AF) phase. Watanabe et al. measured electrical resistivity and magnetic susceptibility for dilute hole-doped La2?x Srx Cu1?y Niy O4 (LSCNO) with x = 0.01.8 A huge increase of resistivity together with a drastic increase of N?el temperature (TN ) was found when doped by a small e amount of Ni. Such the AF order was directly recon?rmed by neutron di?raction using single crystals.9 In addition, the spin structure was found to switch from La2 CuO4 -type to La2 NiO4 -type at y = 0.05, suggesting a change from Ni3+ (S = 1/2) to Ni2+ (S = 1). Machi et al. found that such the Ni-enhanced AF order appears for the SG and the underdoped SC phases too from polycrystalline susceptibility.10 We further explore the Ni-impurity e?ects in the SC phase to study whether the strong hole-localization effect by Ni commonly exists in the entire SC phase. In the present neutron scattering experiments we present the results of Ni-impurity e?ects on the static spin correlations in the SC La2?x Srx CuO4 (LSCO) in the vicinity of the SG-SC boundary. Similar to the result in the SG phase we observed a drastic impurity e?ect in the SC phase. Upon dilute Ni substitution by 3%, both P-SDM and D-SDM considerably shrink in incommensurability (δ), associated with degradation of bulk superconductivity. Subsequent Ni doping induces a bulk three-dimensional AF order with the same spin structure without holes. Based on the hypothesis of the reduction of the e?ective hole concentration by Ni impurity we propose that the previously studied impurity e?ects can be simply interpreted.

address: hiraka@imr.tohoku.ac.jp

2

J. Phys. Soc. Jpn.

Full Paper

Author Name

Fig. 1. Susceptibility data of LSCNO showing Ni-induced spinglass transitions at Tsg ? 5 K.

2. Experimental Single crystals of LSCNO of (x, y) = (0.06, 0.03), (0.06, 0.06), (0.07, 0.03) and (0.07, 0.05) were grown by traveling-solvent-?oating-zone techniques. The crystals have cylindrical shapes of 4 ? 5 mm diameter and ? 20 mm length. After our standard heat treatments under oxygen ?owing gas, those crystals were characterized chemically by ICP measurements and physically using a SQUID magnetometer. In the course of the susceptibility measurements, the volume fraction of bulk susceptibility is found to be strongly suppressed (< 0.01 %) by Ni, judging from the diamagnetic signals. Instead, a spin-glass transition occurs for three crystals (0.06, 0.03), (0.07, 0.03) and (0.07, 0.05) as shown in Fig. 1, but not for the composition of (0.06, 0.06). Neutron scattering experiments were carried out on cold-neutron triple-axis spectrometers LTAS and HER installed in the guide hall of JRR-3 at the Japan Atomic Energy Agency (JAEA), and on SPINS in the research reactor of the National Institute for Standard and Technology in U.S.A. Pyrolytic-graphite (0, 0, 2) re?ection was used in both the monochromator and analyzer. Contaminations of higher-order neutrons were su?ciently suppressed by inserting a Be ?lter into the neutron beam path. Multiple Bragg re?ections were removed by tuning incident energies over the range of 4.5 ? 5 meV. Using pseudo-tetragonal lattice parameters of a? ? 1.66 ??1 (≈ A √ √ ? 2aort ≈ 2b? ) and c? ? 0.48 ??1 , the scattering A ort process was observed in (h, k, 0) and (h, h, l) scattering planes. The horizontal-beam collimation was typically set up to be guide(? 20′ )-80′ -Sample-80′-open(? 180′ ). Some parts of sample-quality check determining twin structures were made on AKANE, a thermal-neutron triple-axis spectrometer of Tohoku University installed at JAEA. All samples studied here consist of twinned crystals, which are naturally caused by the orthorhombic crys-

Fig. 2. Magnetic elastic scattering of La1.94 Sr0.06 Cu1?y Niy O4 around (1/2, 1/2, 0) measured (a,b) along the diagonal direction in (h, k, 0) plane and (c,d) along the l direction in (h, h, l) plane, for (a,c) y = 0.03 and (b,d) y = 0.06. The solid line in (a) is a resolution-convoluted ?t to a simple two-peak cross section along the D-scan, while a single Gaussian curved form is assumed in (b) and (d) for the solid lines. For reference, a resolution-convoluted calculation for y = 0 without P-SDM is shown by a broken line in (a) using D-SDM parameters of ref.6 with an arbitrary intensity scale. Q resolutions are shown by short bars.

tal distortion. The domain distribution was checked by neutron di?raction itself before full measurements of magnetic cross section. Toward the later section of simulation, we remark here that the single crystals of (x, y) = (0.06, 0.03) and (0.07, 0.03) consist of two types of twinning (or four domains) and one type of twinning (or two domains), respectively. In these two samples, the domain population is found to be nearly equal because of the comparable peak intensity from each domain. The orthorhombic lattice parameters are a? ? ort 1.18 ??1 and b? ? 1.17 ??1 in notation of the lowA A ort temperature orthorhombic phase (Bmab), and the orthorhombic distortion does not change by current lightly Ni doping [(b/a) ? 1.01]. For simplicity and convenience, we hereafter use a high-temperature tetragonal notation (I4/mmm) mainly. 3. Results Figures 2(a) and 2(b) show Q spectra around (1/2, 1/2, 0) along the diagonal-scan (D-scan) direction [the inset of Fig. 2(b)] for 3%- and 6%-Ni doped La1.94 Sr0.06 CuO4 , respectively. For reference, a resolution-convoluted spectrum of LSCO with x = 0.06 is shown by a broken line in Fig. 2(a), by using parameters of incommensurate peaks of D-SDM.6 The well-de?ned two-peak structure drastically breaks down upon only

J. Phys. Soc. Jpn.

Full Paper

Author Name

3

Fig. 3. (a) Thermal evolution of magnetic scattering peak at (1/2, 1/2, 0) for LSCNO of (x, y) = (0.06, 0.03) and (0.06, 0.06). Background levels are shown by broken lines. (b) Degradation of TN for Ni-induced N?el ordered state in case of x = y, or e xe? (= x ? y) = 0. Open and closed symbols stand for data from neutron scattering7, 9 and magnetic susceptibility measurements,8, 10 respectively.

Fig. 4. Comparison of Q spectra between the diagonal and parallel scans in La2?x Srx Cu0.97 Ni0.03 O4 for (a) x = 0.06 on LTAS and (b) x = 0.07 on SPINS. The solid lines are ?ts to a two-peak structure along the D-scan. For reference, resolution-convoluted calculations for y = 0 are shown by broken lines with arbitrary intensity scales, which are using parameters of (a) D-SDM of x = 0.066 and (b) P-SDM of x = 0.07,12 respectively. To see easily, an o?set by 100 counts/min is added for the broken line in (b). The Q resolution is shown by short bar.

3%-Ni doping, and a broad commensurate-like peak appears at low temperature [Fig. 2(a)]. Further Ni doping up to 6% induces a commensurate sharp peak, which is resolution-limited and much stronger than that of 3%Ni doped compound [Fig. 2(b)]. Another di?erence of magnetic scattering between the two levels of Ni doping appears in l scans [Figs. 2(c) and 2(d)]. The weak l dependence of the net intensity between T = 7 K and 30 K for y = 0.03 means a weak interlayer coupling of spins. On the other hand, the sharp resolution-limited peak for y = 0.06 corresponds to a bulk AF order in the ground state of La1.94 Sr0.06 Cu0.94 Ni0.06 O4 . Figure 3(a) displays temperature dependences of such the Ni-induced AF order and the magnetic di?use scattering in La1.94 Sr0.06 Cu1?y Niy O4 . TN (? 150 K) of y = 0.06 well follows the x dependence of TN for La2?x Srx Cu1?x Nix O4 7–10 in Fig. 3(b). A linear extrapolation indicates that a bulk AF order can persist up to x ? 0.1. As a preliminary step determining the AF spin structure, we measured three magnetic Bragg re?ections (1, 0, 0)ort , (0, 1, 1)ort and (0, 1, 3)ort of y = 0.06 (not shown). The result shows that the Ni-induced AF spin structure is consistent to that of La2 CuO4 11 with a staggered magnetic moment of 0.2 ? 0.3?B /(Cu site) at base temperature. That is, the AF propagation vector is parallel to [100]ort , while the spins direct [010]ort . In the SC phase of LSCO, particularly near the insulator boundary, both D-SDM and P-SDM coexist at low

temperature.6 In order to clarify Ni-impurity e?ects on each type of SDM, two types of Q scan are carried out in the (h, k, 0) scattering plane. Figure 4 compares Q spectra of the D-scan and P-scan for La2?x Srx Cu0.97 Ni0.03 O4 with x = 0.06 and 0.07. For x = 0.06, the P-scan pro?le is asymmetric about q = 0, possibly due to the domain distribution coupled with a small incommensurability. By contrast, no clear di?erence is observable between the D-scan and P-scan for x = 0.07. Besides, the ?at-toplike cross section suggests a signature of IC correlations remaining. 4. Simulation In order to clarify Ni-impurity e?ects on each type of SDM separately, we focus our discussion on the Q spectra of La2?x Srx Cu0.97 Ni0.03 O4 in Fig. 4. Resolutionconvoluted simulations have been carried out by taking into the experimentally determined orthorhombicdomain distribution. As schematically shown in Fig. 5 and written in the following cross section, we calculate Q spectra around (0.5, 0.5, 0) under the existence of both D-SDM and P-SDM with the incommensurability δd and δp , and the peak width κd and κp , respectively : Isim (Q) ? v100 × Id + Ip + v010 × Id + Ip 2 , (1)

twin

4

J. Phys. Soc. Jpn.

Full Paper

Author Name

Fig. 5. IC-peak con?guration of D-SDM (triangle) and P-SDM (circle), and scan trajectories together with instrumental resolution. IC peaks sprouted only from one (1, 0, 0)ort domain are shown for easy looking. Parameters used in simulation analysis for x = 0.06 in Figs. 6(a)-6(c) are illustrated in real scale. Closed [open] squares and diamonds represent the orthorhombic (1, 0, 0)ort [(0, 1, 0)ort ] position in a four-domain structure caused by two types of twinning. Fig. 6. Two sets of simulation for D-scan and P-scan in La1.94 Sr0.06 Cu0.97 Ni0.03 O4 ; (a)-(c) δd < δp and (d)-(f) δd = δp . Ad : Ap are set to 10 : 0, 9 : 1 and 8 : 2 for (a,d), (b,e) and (c,f), respectively.

2 peaks

2

Id = Ad
Qd

exp ?ln(2) Q ? Qd /κd
2

,

(2)

4 peaks

(3) culated for several sets of δd , δp , Ad and Ap . The total magnetic intensity of each SDM will be closely related to its magnetic-domain volume. Thus, we hereafter evaluate where v100 [v010 ] is the volume fraction of orthorhombic the volume-fraction ratio between D-SDM and P-SDM (1, 0, 0) [ (0, 1, 0) ] domain in a twin. Ad (Ap ) and Qd using Vd : Vp = 3Ad : 8Ap from eqs. (1)-(3). With an (Qp ) represent the IC peak intensity and the peak po- assumption that the spin structure does not change by sition of D-SDM (P-SDM), respectively. The summation 3%-Ni substitution, we can inspect the magnetic-domain with respect to Qd (Qp ) in Id (Ip ) is carried out over two volume before and after Ni doping. The current results (four) IC peaks, since the D-SDM propagates only along of simulation provide us interesting Ni e?ect on the spin the [010]ort direction.4 Basically, Isim consists of two com- correlation of this system though it is di?cult to preponents. One is a contribution from (1, 0, 0)ort domain, cisely determine such the parameters due to the broad and the other from (0, 1, 0)ort domain. For x = 0.06, be- feature of the Q spectra. cause of two types of twin formation in the sample, we Figure 6 shows typical examples of simulation for x = repeat this addition for another twin also. As for the 0.06. In order to reproduce the observed asymmetry in prefactor (1/2) of Id in (0, 1, 0)ort domain of eq. (1), the P-scan [Fig. 4(a)], (1) δd should be quite di?erent we referred the experimental result for SG LSCO with from δp (δd = 0.02 r.l.u. < δp = 0.04 r.l.u.) and (2) x = 0.05.4 Ad : Ap ? 9 : 1, as shown in Fig. 6(b). This intensity Because of the many degrees of freedom in the above ratio corresponds to Vd : Vp ? 3 : 1. These parameters model cross section, a proper initial set of SDM param- are substantially di?erent from results of Ni-free sample eters is much required to converge our simulation study. with x = 0.06; δd ≈ δp ? 0.05 r.l.u. and Vd : Vp ? 2 : 1.6 To get such the parameters, as a ?rst step, a simple For x = 0.07 the calculated Q spectra are less sensianalysis was preparatively done by assuming a two-peak structure along the D-scan direction without P-SDM. A fair agreement is shown in Fig. 4 by curved lines, and the resultant δd and κd are listed in Table I. Note that the Table I. Referenced data for coming simulation study, estimated through a simple two-peak ?t. For reference, data without Ni incommensurability de?nitely decreases with Ni whereas are also shown. Note that only the data of x = 0.07 come from the peak width does not change so much in this prelimiP-SDM. nary analysis. x y δd (r.l.u.) κd (??1 ) κd (r.l.u.) A To simplify the simulation, we ?x the peak width to 0.06 0.03 0.017(3) 0.030(4) 0.018(2) present κd = 0.03 ??1 for both x = 0.06 and 0.07, based on A 0.07 0.03 0.029(1) 0.037(4) 0.022(2) present Table I and on the data near the SG-SC boundary in 0.06 0 0.053(2) 0.039(4) 0.023(2) ref.6 ? ? 6 0.07 0 ? 0.069 ? 0.037 ? 0.022 ref.12 LSCO. Besides, we assume κd = κp according to the result of x = 0.06 without Ni.6 Hence, Isim was calIp = Ap
Qp

exp ?ln(2) Q ? Qp /κp

.

J. Phys. Soc. Jpn.

Full Paper

Author Name

5

Fig. 7. Two sets of simulation for D-scan and P-scan in La1.93 Sr0.07 Cu0.97 Ni0.03 O4 ; (a)-(c) δd < δp and (d)-(f) δd = δp . Ad : Ap are set to 10 : 0, 8 : 2 and 6 : 4 for (a,d), (b,e) and (c,f), respectively. FWHM of the ?at-top cross section in the D-sacn is shown for (b) and (f).

tive to the parameters compared to the case of x = 0.06. Nonetheless some detailed Ni e?ect was obtained to explain the speci?c features seen in Fig. 4(b); namely, the coincidence of two types of scans and the ?at-top-like pro?les. In this sense, Figs. 7(b) and 7(f) are good candiates to reproduce the experimental data of Fig. 4(b). Further, due to the observed FWHM (? 0.18 ??1 ) of the A ?at-top cross section along the D-scan, Fig. 7(b) looks better than Fig. 7(f). Therefore, (1) δd is slightly smaller than δp (δd = 0.03 r.l.u. < δp = 0.04 r.l.u.) and (2) Ad : Ap ? 8 : 2. As a conclusion, δp is much smaller than that of Ni-free sample with x = 0.07 (δp ? 0.07 r.l.u.12 ), and the Vd ratio (Vd : Vp ? 1.5 : 1) is quite large, compared to the Ni-free case (Vd : Vp ? 0.7 : 1).13 5. Discussion We studied Ni-impurity e?ect on spin correlation in the SC phase near the SG-SC boundary. Similar to the previous results in the AF ordered phase9 and SG phase,7 Ni substitution drastically changes the spin correlation. All these facts observed in the wide hole-doping range are commonly explained with an intuitive scenario that Ni and hole couples strongly on the CuO2 planes, thus reducing the number of mobile holes. Indeed, Fig. 7 supports this hypothesis. Surprisingly, the δd as well as the Tela in Ni-doped LSCO well follow the data in Ni-free LSCO,3–6 when the e?ective hole concentration is supposed to be xe? = x ? y. In the present case, since the P-SDM and the D-SDM coexist at low temperature, we can further discuss separately the e?ect of Ni impurity on each SDM phase. The results of analysis taking into the domain distribution

Fig. 8. Plots of (a) incommensurability δd and (b) onset temperature Tela of D-SDM peaks against xe? (= x ? y) in LSCNO. Filled circles and squares represent data from the SC phase and the SG phase,7 respectively. The data of Ni-free LSCO3–6 are shown by open circles. Tsg determined by susceptibility is also plotted in (b).

provide us interesting information on the spin correlation near the boundary between the SG and SC phases. The previous study without Ni impurities demonstrates a sudden appearance of P-SDM with δp = 0.049 (r.l.u in tetragonal unit) upon entering the SC phase with x > xcri = 0.055, suggesting a ?rst order transition between the D-SDM and P-SDM.6 Since the maximum value of Tc at a given doping x is proportional to the δp , the ?rst-order transition suggests the existence of a ?nite minimum value of Tc . Through simulation analysis for the two types of Q scans in Figs. 6 and 7, speci?c features of the SDM are newly found. That is, the volume fraction of D-SDM increases by Ni doping, and the incommensurability tends to give a discrepancy between D-SDM and P-SDM. Intuitively, the expanding Vd is considered as a result of switch from P-SDM to D-SDM at a low temperature by reducing xe? down to bellow xcri . The survival of partial P-SDM for xe? < xcri may occur when holes distribute inhomogeneously in CuO2 plane and when the hole density exceeds beyond xcri in local. Such the microscopic inhomogeneity should be introduced most likely by Ni impurity, and further by supercooling the mobile holes because of crossing boundary of the ?rst-order transition at xcri . The slightly lower values of δd for Ni-doped samples compared to those of Ni-free samples in Fig. 8(a) as well as the simulation result of δd < δp supports this scenario.

6

J. Phys. Soc. Jpn.

Full Paper

Author Name

It is noteworthy that this hole-localization scenario around Ni reasonably explains Ni-impurity e?ects reported previously in other kinds of physical quantities. We examples three cases below. (i) Neutron resonance peak; The well-known magnetic resonant mode appears in inelastic neutron scattering for SC YBa2 Cu3 O6+x with resonance energy Er .14–16 With the help of a relation between Tc and the hole concentration (p),17 Er can be described as a function of p. In YBa2 (Cu0.97 Ni0.03 )3 O7 , the Er decreases from Ni-free 41 meV to 35 meV.18 This reduction corresponds to ?p ? ?0.05 and it agrees semiquantitatively with the doped Ni-concentration 0.03. (ii) Pseudo-gap; Ni impurity enhances the normal-state pseudo-gap in the c-axis optical conductivity of underdoped (Sm,Nd)Ba2 (Cu1?y Niy )3 O7?δ .19 The increasing pseudo-gap is considered as a natural consequence of the underdoping by Ni. (iii) STM; The SC-coherence peak is little a?ected by Ni in Bi2 Sr2 CaCu2 O8+δ .20 A strong coupling of Ni and hole produces a spin state of S = 1/2 (either 3d7 with Ni3+ , or 3d8 L), and then gives a minimum perturbation to the underlying spin-1/2 framework. The superconductivity, therefore, will be less damaged by Ni. Finally, recent our XAFS experiments using synchrotron radiation support our scenario.21 That is, the valence states of Ni in La1.94 Sr0.06 Cu0.97 Ni0.03 O4 and La1.94 Sr0.06 Cu0.94 Ni0.06 O4 are much di?erent from a Ni2+ state, thus indicating most probably either a Ni3+ or a strongly hole-bound Ni2+ state. Acknowledgment We are grateful to J. Ho for neutron scattering on SPINS at NIST, to S.-H. Lee, M. H¨ cker, M. Kofu, u M. Fujita, Y. Itoh, W. Koshibae, K. Tsutsui, T. Tohoyama and M. Ogata for stimulating discussions. We also thank K. Nemoto and N. Aso for neutron scattering on AKANE and HER at JAERI, and M. Sakurai for growing the single crystals. The work at Tohoku University was supported by grants from the Ministry of Education, Culture, Sports, Science and Technology. This study was supported by the U.S.-Japan Cooperative Neutron-Scattering Program. Financial support from the U.S. Department of Energy under Contract DE-AC0298CH10886 is also gratefully acknowledged. Work at SPINS is based upon activities supported by the NSF under DMR-9986442. We also acknowledge the U.S. Dept. of Commerce, NIST Center for Neutron Research, for providing the neutron scattering facilities used in this study.

1) K. Yamada, C.H. Lee, K. Kurahashi, J. Wada, S. Wakimoto, S. Ueki, H. Kimura, Y. Endoh, S. Hosoya, G. Shirane, R.J. Birgeneau, M. Greven, M.A. Kastner, and Y.J. Kim: Phys. Rev. B 57 (1998) 6165. 2) S. Wakimoto, H. Zhang, K. Yamada, I. Swainson, Hyunkyung Kim, and R.J. Birgeneau: Phys. Rev. Lett. 92 (2004) 217004. 3) S. Wakimoto, G. Shirane, Y. Endoh, K. Hirota, S. Ueki, K. Yamada, R. J. Birgeneau, M. A. Kastner, Y. S. Lee, P. M. Gehring, and S.-H. Lee: Phys. Rev. B 60 (1999) R769. 4) S. Wakimoto, R. J. Birgeneau, M. A. Kastner, Y. S. Lee, R. Erwin, P. M. Gehring, and S.-H. Lee, M. Fujita, K. Yamada, Y. Endoh, K. Hirota, and G. Shirane: Phys. Rev. B 61 (2000) 3699. M. Matsuda, M. Fujita, K. Yamada, R.J. Birgeneau, M.A. Kastner, H. Hiraka, Y. Endoh, S. Wakimoto, and G. Shirane: Phys. Rev. B 62 (2000) 9148. M. Fujita, K. Yamada, H. Hiraka, P.M. Gehring, S.-H. Lee, S. Wakimoto, and G. Shirane: Phys. Rev. B 65 (2002) 064505. M. Matsuda, M. Fujita, and K. Yamada: Phys. Rev. B 73 (2006) 140503(R). N. Watanabe, T. Masui, Y. Itoh, T. Machi, I. Kato, N. Koshizuka, and M. Murakami: Physica C 388-389 (2003) 241. H. Hiraka, T. Machi, N. Watanabe, Y. Itoh, M. Matsuda, and K. Yamada: J. Phys. Soc. Jpn. 74 (2005) 2197. T. Machi, I. Kato, R. Hareyama, N. Watanabe, Y. Itoh, N. Koshizuka, S. Arai, and M. Murakami: Physica C 388389 (2003) 233. D. Vaknin, S.K. Sinha, D.E. Moncton, D.C. Johnston, J.M. Newsam, C.R. Sa?nya, and H.E. King Jr.: Phys. Rev. Lett. 58 (1987) 2802. H. Hiraka, Y. Endoh, M. Fujita, Y.S. Lee, J. Kulda, A. Ivanov, and R.J. Birgeneau: J. Phys. Soc. Jpn. 70 (2001) 853. This ratio is obtained from elastic scattering measurements using thermal neutron beams (energy resolution ? 2 meV), and it is qualitatively consistent with the description about x = 0.07 in ref.,6 ”... parallel component dominates the diagonal one.” J. Rossat-Mignod, L.P. Regnault, C. Vettier, P. Bourges, P. Burlet, J. Bossy, J.Y. Henry, and G. Lapertot: Physica C 185 (1991) 86. H.F. Fong, B. Keimer, D.L. Milius, and I.A. Aksay: Phys. Rev. Lett. 78 (1997) 713. P. Dai, H.A. Mook, R.D. Hunt, and F. Dogan: Phys. Rev. B 63 (2001) 054525. For example, J.L. Tallon, C. Bernhard, H. Shaked, R.L. Hitterman, and J.D. Jorgensen: Phys. Rev. B 51 (1995) 12911. Y. Sidis, P. Bouges, H.F. Fong, B. Keimer, L.P. Legnault, J. Bossy, A. Ivanov, B. Hennion, P. Gautier-Picard, G. Collin, D.L. Millius, and I.A. Aksay: Phys. Rev. Lett. 84 (2000) 5900. A.V. Pimenov, A.V. Boris, Li Yu, V. Hinkov, Th. Wolf, J.L. Tallon, B. Keimer, and C. Bernhard: Phys. Rev. Lett. 94 (2005) 227003. E.W. Hudson, K.M. Lang, V. Madhavan, S.H. Pan, H. Eisaki, S. Uchida, and J.C. Davis: Nature 411 (2001) 920. H. Hiraka et al., in preparation for publication.

5)

6) 7) 8)

9) 10)

11)

12) 13)

14)

15) 16) 17) 18)

19)

20) 21)



★相关文章:
热文推荐
猜你喜欢
友情链接: 医学资料大全 农林牧渔 幼儿教育心得 小学教育 中学 高中 职业教育 成人教育 大学资料 求职职场 职场文档 总结汇报