betaFeedback

WorksPhysics and Astronomy

Spin-lattice-coupled helical magnetic order in breathing pyrochlore magnets, CuAlCr4_{4}S8_{8} and CuGaCr4_{4}S8_{8}

Abstract

We report low-temperature powder X-ray and neutron diffraction studies on breathing pyrochlore magnets CuMMCr4_{4}S8_{8} (MM = Al, Ga), which undergo a magnetic transition at TN≈T_{\rm N} \approx 21 and 31 K for {MM = Al and Ga, respectively. X-ray diffraction reveals that the magnetic transition accompanies a structural transition from cubic F4‾3mF{\overline 4}3m to polar orthorhombic Imm2Imm2 symmetry for both the compounds, with larger distortion observed for MM = Ga at low temperatures. Neutron diffraction reveals incommensurate magnetic modulation Q=(qIC,0.5,0){\mathbf Q} = (q_{\rm IC}, 0.5, 0) in the orthorhombic setting, where qIC≈q_{\rm IC} \approx 0.39 and 0.31 for MM = Al and Ga, respectively. Our magnetic-structure analysis suggests cycloid-type magnetic order but not proper-screw type for both the compounds. We find strong correlation between the local spin configuration and Cr-Cr bond lengths, indicating that the spin-lattice coupling as well as the magnetic frustration play an important role in determining the ground state. CuMMCr4_{4}S8_{8} potentially offers a platform to explore magnetoelectric effects arising from the helimagnet driven electric polarity.

Introduction

The breathing pyrochlore magnet, consisting of a three-dimensional network of alternately arranged small and large tetrahedra, has recently been recognized as a new class of frustrated spin systems.1) The introduction of inequivalent nearest-neighbor (NN) exchange interactions JJ and J′J' in the small and large tetrahedra, respectively, provides another degree of freedom absent in the regular pyrochlore lattice. The emergence of unconventional ground states due to the breathing nature, such as the tetramer-singlet state,2) commensurate hedgehog-antihedgehog lattice,3) and spin-lattice-coupled magnetic superstructures4) has been proposed.

A-site ordered chromium thiospinels with the Cr3+^{3+} ion carrying an S=3/2S = 3/2 spin offer a fertile playground for systematically investigating the magnetism of the breathing pyrochlore classical Heisenberg models with various combinations of JJ and J′J'.5–14) Density functional theory (DFT) calculations predict that both JJ and J′J' are ferromagnetic (FM) in LiMCr₄S₈ (MM = Ga, In), whereas JJ is antiferromagnetic (AFM) and J′J' is FM in CuMCr₄S₈ (MM = Ga, In).9,14) If the FM J′J' is dominant in the latter case, four spins in each large tetrahedron may behave as an effective S=6S = 6 spin, which is observed as a FM cluster excitation in CuInCr₄S₈.11) Relatively strong second-NN and third-NN AFM exchange interactions are also present in Cr thiospinels,9,14) giving rise to the magnetic frustration pervading the entire lattice.

One important character of Cr spinels is a predominant role of the spin-lattice coupling due to the distance-sensitive direct exchange interaction between NN spins. At low temperatures, Cr spinels undergo a magnetic transition accompanied by crystallographic symmetry breaking.16–24) Table I summarizes the crystal structure and magnetic propagation vector in low-temperature phases of several related compounds, revealed by previous X-ray diffraction (XRD) and neutron diffraction (ND) experiments. The change in the non-magnetic AA-sites results in a variety of magnetic states with different crystallographic symmetry. A phase separation is observed in some compounds.21–24) These observations suggest that the ground state of the Cr spinel is selected by a delicate balance between the exchange and elastic energy, and also significantly influenced by sample quality and strain.

In this study, we focus on the crystal and magnetic structures of the breathing pyrochlore magnets CuMCr₄S₈ (MM = Al, Ga) at low temperatures. Both compounds possess a cubic crystal structure with the F4ˉ3mF\bar{4}3m space group at room temperature.12,14) Upon cooling, a magnetic transition occurs at TN≈21T_{\mathrm{N}} \approx21 K for MM = Al12) and TN≈31T_{\mathrm{N}} \approx31 K for MM = Ga,5,14) accompanied by a sudden drop in magnetic susceptibility. The crystal and magnetic structures in the low-temperature phase have not been fully elucidated so far in spite of a few reports of X-ray and neutron studies.6,12,14) We reveal through powder XRD experiments that CuAlCr₄S₈ and CuGaCr₄S₈ commonly undergo a transition to an orthorhombic structure with the Imm2Imm2 space group at TNT_{\mathrm{N}}. Furthermore, powder ND experiments reveal the emergence of cycloidal order with an incommensurate propagation vector of Q=(qIC,0.5,0)o\mathbf{Q} = (q_{\mathrm{IC}}, 0.5, 0)_{\mathrm{o}} in the orthorhombic setting, where qIC≈0.39q_{\mathrm{IC}} \approx0.39 and 0.31 for MM = Al and Ga, respectively. The identified ground states are unique to the present compounds among the Cr spinel families (see Table I), which can be attributed to the breathing degree of freedom in addition to the magnetic frustration and spin-lattice coupling.

Method

Polycrystalline samples of CuAlCr₄S₈ and CuGaCr₄S₈ were synthesized by the conventional solid-state reaction method. A stoichiometric mixture of copper (99.99%), aluminum (99.99%)/gallium (99.999%), chromium (99.99%), and sulfur (99.99%) powders was sealed in an evacuated quartz tube and heated at 900°C for 48 h. Subsequently, the sintering process was repeated twice at 1000°C for 96 h. Magnetization was measured using a superconducting quantum interference device (MPMS, Quantum Design). Thermal expansion was measured by the fiber-Bragg-grating (FBG) method using an optical sensing instrument (Hyperion si155, LUNA) in a cryostat equipped with a superconducting magnet (Spectromag, Oxford). The optical fiber was glued to the sintered samples using Stycast1266.

Crystal structureMagnetic propagation vectorTemperatureReference
MgCr2_2O4_4Tetragonal I41/amdI4_1/amd (c/a=0.9979c/a = 0.9979)Q1=(0.5,0.5,0)+Q2=(1,0,0.5)\mathbf{Q}_1 = (0.5, 0.5, 0) + \mathbf{Q}_2 = (1, 0, 0.5)T<TN=12.4T < T_{\mathrm{N}} = 12.4 KRefs. 16, 17
ZnCr2_2O4_4Tetragonal Iˉ4m2\bar{I}4m2 (c/a=0.9982c/a = 0.9982)Q1=(0.5,0.5,0)\mathbf{Q}_1 = (0.5, 0.5, 0) & Q2=(1,0,0.5)\mathbf{Q}_2 = (1, 0, 0.5)T<TN=12.5T < T_{\mathrm{N}} = 12.5 KRef. 18
CdCr2_2O4_4Tetragonal I41/amdI4_1/amd (c/a=1.0043c/a = 1.0043)Q=(0,0.09,1)\mathbf{Q} = (0, 0.09, 1)T<TN=7.8T < T_{\mathrm{N}} = 7.8 KRef. 19
HgCr2_2O4_4Orthorhombic FdddFdddQ1=(0.5,0,1)\mathbf{Q}_1 = (0.5, 0, 1) & Q2=(1,0,0)\mathbf{Q}_2 = (1, 0, 0)T<TN=5.8T < T_{\mathrm{N}} = 5.8 KRefs. 20, 21
LiGaCr4_4O8_8Tetragonal Iˉ4m2\bar{I}4m2 (c/a=1.0056c/a = 1.0056) & Cubic F4ˉ3mF\bar{4}3mQtetra=(1,0,0.5)\mathbf{Q}_{\mathrm{tetra}} = (1, 0, 0.5) & Qcubic=(1,0,0)\mathbf{Q}_{\mathrm{cubic}} = (1, 0, 0)T<TN=13∼16T < T_{\mathrm{N}} = 13 \sim16 KRef. 22
LiInCr4_4O8_8Tetragonal Iˉ4m2\bar{I}4m2 (c/a=0.9910c/a = 0.9910) & Cubic F4ˉ3mF\bar{4}3mQtetra=(1,0,0.2)\mathbf{Q}_{\mathrm{tetra}} = (1, 0, 0.2) & Qcubic=(1,0,0)\mathbf{Q}_{\mathrm{cubic}} = (1, 0, 0)T<TN=14.5T < T_{\mathrm{N}} = 14.5 KRef. 23
ZnCr2_2S4_4Tetragonal I41/amdI4_1/amd (c/a=0.9985c/a = 0.9985) Orthorhombic ImmaImma (c/a=1.0003c/a = 1.0003, γ=89.832∘\gamma= 89.832^\circ)Q1=(0,0,0.79)\mathbf{Q}_1 = (0, 0, 0.79) & Q2=(0.5,0.5,0)\mathbf{Q}_2 = (0.5, 0.5, 0) Q=(1,0.5,0)\mathbf{Q} = (1, 0.5, 0)TN2<T<TN1=15T_{\mathrm{N}2} < T < T_{\mathrm{N}1} = 15 K T<TN2=8T < T_{\mathrm{N}2} = 8 KRef. 24
CuAlCr4_4S8_8Orthorhombic Imm2Imm2 (c/a=1.0004c/a = 1.0004, γ=89.833∘\gamma= 89.833^\circ)Q=(0.89,0.11,0)\mathbf{Q} = (0.89, 0.11, 0) [Q=(0.39,0.5,0)o\mathbf{Q} = (0.39, 0.5, 0)_o]T<TN=21T < T_{\mathrm{N}} = 21 KThis work
CuGaCr4_4S8_8Orthorhombic Imm2Imm2 (c/a=1.0005c/a = 1.0005, γ=89.743∘\gamma= 89.743^\circ)Q=(0.81,0.19,0)\mathbf{Q} = (0.81, 0.19, 0) [Q=(0.31,0.5,0)o\mathbf{Q} = (0.31, 0.5, 0)_o]T<TN=31T < T_{\mathrm{N}} = 31 KThis work

Table 1. Crystal structure and magnetic propagation vector of a magnetic ordering phase in various chromium spinel compounds. For LiGaCr4_4O8_8 and LiGaCr4_4O8_8, a phase separation of tetragonal and cubic phases was reported. The magnetic propagation vectors are based on the original cubic cell. “+” indicates a superposition of multiple modulations, whereas “&” indicates a phase separation.

Powder XRD experiments were performed using a Rigaku SmartLab diffractometer at Materials Characterization Support Team, RIKEN CEMS. The incident X-ray beam was monochromatized to Cu-Kα1K\alpha_1 radiation by a Johansson-type monochromator with a Ge(111) crystal. The Rietveld analysis was performed using the RIETAN software.25) Crystal structures were visualized by using the VESTA software.26) Powder ND experiments were performed at a triple-axis spectrometer PONTA(5G) installed in JRR-3 of the Japan Atomic Energy Agency. The spectrometer was operated in the two-axis mode, and the horizontal collimation was 40′-40′-40′-40′. An incident neutron beam with the wavelength of 2.392 Å was obtained by a PG (002) monochromator.

Results

Magnetization and thermal expansion

Figure 1 shows the temperature dependence of magnetic susceptibility M/HM/H measured at 1 T and thermal expansion ΔL/L\Delta L/L at 0 T for CuAlCr4_4S8_8 (M=AlM = \mathrm{Al}, Ga). The M/HM/H–TT curve of CuAlCr4_4S8_8 well reproduces the data reported in Ref. 12. For both the compounds, M/HM/H suddenly drops at TNT_{\mathrm{N}}, suggesting the development of AFM long-range order in a low-temperature phase. This phase transition accompanies a hysteresis [see insets of Figure 1(a) and Figure 1(c)], indicative of a first-order transition. A steep decrease in the sample length can be seen above TNT_{\mathrm{N}} in the ΔL/L\Delta L/L–TT curves, suggesting the occurrence of a structural transition associated with volume reduction. As shown below, similar behaviors in M/HM/H and ΔL/L\Delta L/L between CuAlCr4_4S8_8 and CuGaCr4_4S8_8 can be ascribed to the appearance of similar spin-lattice-coupled ground states.

Figure 1. Temperature dependence of magnetic susceptibility M/HM/H at 1 T (top) and thermal expansion ΔL/L\Delta L/L at 0 T (bottom) for CuAlCr4_4S8_8 (left) and CuGaCr4_4S8_8 (right). The data of CuGaCr4_4S8_8 are excerpt from Ref. 14. Magnetization was measured during a cooling process down to 2 K and a subsequent warming process. The inset in (a) and (c) displays an enlarged view of the M/HM/H–TT curve in the vicinity of the magnetic transition temperature TNT_{\mathrm{N}}.

Powder X-ray diffraction experiments

Figure 2(a)–Figure 2(c) and Figure 2(d)–Figure 2(f) show changes in XRD peak profiles of selected Bragg reflections across TNT_{\mathrm{N}} for CuAlCr4_4S8_8 and CuGaCr4_4S8_8, respectively. Here, the indices are based on the original cubic cell. We observe a common feature in the peak splitting patterns for the two compounds: (i) the hhhhhh reflection splits into two peaks with an intensity ratio of approximately 1:1, (ii) the hh0hh0 reflection splits into three peaks with an intensity ratio of approximately 1:4:1, and (iii) the h00h00 reflection does not exhibit clear splitting. These observations are compatible with orthorhombic distortion characterized by a=b≠ca = b \ne c, α=β=γ=90∘\alpha= \beta= \gamma= 90^\circ, from the original cubic FF lattice, resulting in the orthorhombic II lattice with a halved unit cell, as depicted in Figure 3(a). Since no additional Bragg peaks appear, we identify the crystallographic space group below TNT_{\mathrm{N}} as Imm2Imm2 (No. 44). The split-tings widths of the hhhhhh and hh0hh0 peaks, which reflect the deviation of the angle γ\gamma from 90∘90^\circ, are larger for CuGaCr4_4S8_8 than for CuAlCr4_4S8_8 [Figure 2(a), Figure 2(c), Figure 2(d), and Figure 2(f)]. This indicates that CuGaCr4_4S8_8 exhibits larger rhombic distortion in the abab plane compared to CuAlCr4_4S8_8. A structural transition from F4ˉ3mF\bar{4}3m to Imm2Imm2 symmetry has been reported in lacunar spinels GeV4_4S8_827) and GaMo4_4Se8_828), where the orbital degree of freedom plays an important role. On the other hand, only the in-

Figure 2. Results of powder XRD experiments on CuMCr4_4S8_8 (MM = Al, Ga) at low temperatures. [(a)–(f)] Experimental XRD profiles at selected Bragg positions at 25 K/35 K (gray) and 4 K (red) for [(a)–(c)] MM = Al and [(d)–(f)] MM = Ga, respectively. The indices, 800, 444, and 880, are defined based on the original cubic cell. The observed peak splittings are compatible with the orthorhombic Imm2Imm2 space group (No. 44). [(g)(h)] Experimental XRD patterns at 4 K (red open circles) and the calculated ones obtained by the Rietveld analysis assuming the Imm2Imm2 space group (black lines) for (g) MM = Al and (h) MM = Ga. Blue vertical bars indicate positions of the Bragg reflections, and the green line is a difference between the experimental and calculated patterns. Bragg peaks originating from the copper plate are removed from the analysis.

Summary of crystallographic properties in CuMCr4S8

Figure 3. Summary of crystallographic properties in CuMCr4_4S8_8 (MM = Al, Ga). (a) Schematic representation of the lattice distortion in the abab plane across TNT_{\mathrm{N}}. (b) Crystal structures within a unit cell for the F‾43m\overline{F}43m (left) and Imm2Imm2 (right) space groups. For the latter, the crystallographic polarity (PcP_{\mathrm{c}}) exists along the cc axis. (c) Cr tetrahedral network and the definition of the NN Cr–Cr bond lengths in the F‾43m\overline{F}43m (left) and Imm2Imm2 (right) structures. Bonds of the same length are drawn in the same color and thickness. [(d)–(k)] Temperature dependence of (d)(h) lattice constants γ\gamma, (e)(i) aa and cc, and (f)(j) volume VV of the original cubic cell, and (g)(k) the NN Cr–Cr bond lengths for MM = Al (left) and MM = Ga (right).

terplay between spin and lattice degrees of freedom should be a driving force of the structural transition in CuMCr4_4S8_8.

On the basis of the above-mentioned considerations, we perform a Rietveld analysis of the XRD patterns at low temperatures. A satisfactory fit is achieved by assuming the Fˉ43m\bar{F}43m space group above TNT_{\mathrm{N}} and the Imm2Imm2 space group below TNT_{\mathrm{N}}, as shown in Figure 2(g) and Figure 2(h). Tables II and III summarize the obtained structural parameters and reliability factors for CuAlCr4_4S8_8 and CuGaCr4_4S8_8, respectively. As depicted in the right panel of Figure 3(a), the lattice constants of the orthorhombic cell are defined such that a′>b′a' > b', where the a′a' and b′b' axes are set by rotating the cubic aa and bb axes by 45∘45^\circ. The crystal structure within the corresponding unit cell for each space group is schematically illustrated in Figure 3(b). In the Imm2Imm2 structure, there exist two types of Cr sites, Cr1 and Cr2, alternately stacking along the cc axis. Notably, the averaged zz coordinates of the Cr and S atoms deviate from z=0.5z = 0.5 below TNT_{\mathrm{N}}, leading to a polar structure along the cc axis. The rhombic distortion results in three types of NN Cr–Cr bonds within each tetrahedron, as shown in the right panel of Figure 3(c). In the following, we define the Cr–Cr bond lengths along the a′a', b′b', and four equivalent ⟨111⟩\langle111\rangle axes in the orthorhombic setting below TNT_{\mathrm{N}} as dad_a (da′d'_a), dbd_b (db′d'_b), and d111d_{111} (d111′d'_{111}) in each small (large) tetrahedron. Figures 3(d)–3(k) show the temperature dependence of various crystallographic parameters for CuAlCr4_4S8_8 and

AtomSiteOcc.xxyyzzBB (Å2^2)
Cu4a4a10000.380(54)
Al4d4d10.750.750.750.015(93)
Cr16e16e10.37008(12)xxxx0.391(19)
S116e16e10.13303(13)xxxx0.2 (fix)
S216e16e10.61671(10)xxxx0.2 (fix)
Cu2a2a10.50.500.426(66)
Al2b2b100.50.750.063(118)
Cr14c4c10.50.24442(83)0.63480(80)0.321(60)
Cr24d4d10.25181(100)00.86890(89)0.174(55)
S14c4c100.27483(85)0.36926(90)0.2 (fix)
S24c4c100.24084(79)0.88640(77)0.2 (fix)
S34d4d10.23999(98)00.11376(77)0.2 (fix)
S44d4d10.26817(105)00.63111(86)0.2 (fix)

Table 2. Structural parameters of CuAlCr4_4S8_8 obtained from the Rietveld analysis at 25 K (top) and 4 K (bottom) based on the Fˉ43m\bar{F}43m (No. 216) and Imm2Imm2 space group (No. 44), respectively. The lattice constants are a=9.90259(35)a = 9.90259(35) Å at 25 K, and a′=7.00809(37)a' = 7.00809(37) Å, b′=6.98774(37)b' = 6.98774(37) Å, c=9.90018(53)c = 9.90018(53) Å at 4 K. T=25T = 25 K (Rwp=4.82R_{\mathrm{wp}} = 4.82, Rp=2.77R_{\mathrm{p}} = 2.77, Re=2.60R_{\mathrm{e}} = 2.60, S=1.85S = 1.85). T=4T = 4 K (Rwp=5.13R_{\mathrm{wp}} = 5.13, Rp=2.99R_{\mathrm{p}} = 2.99, Re=2.61R_{\mathrm{e}} = 2.61, S=1.97S = 1.97).

AtomSiteOcc.xxyyzzBB (Å2^2)
Cu4a4a10000.720(106)
Ga4d4d10.750.750.750.087(70)
Cr16e16e10.36999(13)xxxx0.214(17)
S116e16e10.13124(15)xxxx0.2 (fix)
S216e16e10.61530(17)xxxx0.2 (fix)
Cu2a2a10.50.500.157(102)
Ga2b2b100.50.750.454(84)
Cr14c4c10.50.23763(64)0.63302(66)0.040(44)
Cr24d4d10.25482(105)00.87053(88)0.344(49)
S14c4c100.26689(83)0.37387(93)0.2 (fix)
S24c4c100.23284(71)0.89064(93)0.2 (fix)
S34d4d10.23080(107)00.11709(82)0.2 (fix)
S44d4d10.26426(115)00.63118(95)0.2 (fix)

Table 3. Structural parameters of CuGaCr4_4S8_8 obtained from the Rietveld analysis at 35 K (top) and 4 K (bottom) based on the Fˉ43m\bar{F}43m (No. 216) and Imm2Imm2 space group (No. 44), respectively. The lattice constants are a=9.90691(48)a = 9.90691(48) Å at 35 K, and a′=7.01696(48)a' = 7.01696(48) Å, b′=6.98556(47)b' = 6.98556(47) Å, c=9.90588(67)c = 9.90588(67) Å at 4 K. T=35T = 35 K (Rwp=3.36R_{\mathrm{wp}} = 3.36, Rp=2.24R_{\mathrm{p}} = 2.24, Re=2.90R_{\mathrm{e}} = 2.90, S=1.16S = 1.16). T=4T = 4 K (Rwp=2.85R_{\mathrm{wp}} = 2.85, Rp=1.79R_{\mathrm{p}} = 1.79, Re=2.90R_{\mathrm{e}} = 2.90, S=0.97S = 0.97).

CuGaCr4_4S8_8. The two compounds exhibit comparable volume reduction, ΔV/V≈1.2×10−3\Delta V/V \approx1.2 \times10^{-3}, across TNT_{\mathrm{N}} [Figure 3(f) and Figure 3(j)]. Meanwhile, the lattice distortion is more significant in CuGaCr4_4S8_8 compared to CuAlCr4_4S8_8: c/a=1.0004c/a = 1.0004 and γ=89.833∘\gamma= 89.833^\circ for CuAlCr4_4S8_8, and c/a=1.00005c/a = 1.00005 and γ=89.743∘\gamma= 89.743^\circ for CuGaCr4_4S8_8 at 4 K [Figure 3(d), Figure 3(e), Figure 3(h), and Figure 3(i)]. Above TNT_{\mathrm{N}}, the ratios of the two kinds of NN Cr–Cr bond lengths, dd and d′d', are nearly identical between the two compounds: d′/d=1.082(3)d'/d = 1.082(3) for CuAlCr4_4S8_8 at 25 K and d′/d=1.084(3)d'/d = 1.084(3) for CuGaCr4_4S8_8 at 35 K. Below TNT_{\mathrm{N}}, however, the difference between dad_a and da′d'_a, as well as between dbd_b and db′d'_b, becomes much smaller for CuAlCr4_4S8_8 than for CuGaCr4_4S8_8 [Figure 3(g) and Figure 3(k)]. In other words, the breathing of the pyrochlore network is greatly suppressed below $T_{\mathrm{N}} for CuAlCr4_4S8_8. The resulting volume ratio of the larger tetrahedron to the smaller one below TNT_{\mathrm{N}} is 1.21 for CuAlCr4_4S8_8 and 1.27 for CuGaCr4_4S8_8.

Powder neutron diffraction experiments

Figure 4(a) and Figure 4(b) show the experimental powder ND patterns for MM = Al and Ga, respectively. Here, the splitting is not observed for any nuclear Bragg peaks at 2 K due to the rather low QQ resolution of the instrument. Figures. Figure 4(c) and Figure 4(d) compare the observed intensities IobsI_{\mathrm{obs}} of nuclear Bragg peaks after the Lorentz factor correction with the calculated ones IcalI_{\mathrm{cal}} based on the 4-K crystal structure revealed by the XRD (Tables II and III). A least-squares fit confirms good agreement between IobsI_{\mathrm{obs}} and IcalI_{\mathrm{cal}} for both the compounds; the intensity reliability factor RBR_{\mathrm{B}} defined as

Experimental powder neutron diffraction patterns and magnetic modulation

Figure 4. [(a)(b)] Experimental powder ND patterns at 40 K/50 K (gray) and 2 K (blue) for (a) CuAlCr4_4S8_8 and (b) CuGaCr4_4S8_8. The strong peak at 2θ=61∘2\theta= 61^\circ originates from aluminum. The indices of the nuclear Bragg peaks are defined in the cubic setting. For CuGaCr4_4S8_8, the 200 and 420 peaks are not observed due to the close scattering length of Cu and Ga atoms. Black and red vertical bars indicate positions of the nuclear and magnetic Bragg reflections, respectively. [(c)(d)] IobsI_{\mathrm{obs}} versus IcalI_{\mathrm{cal}} plot of the nuclear Bragg peaks at 2 K for (c) CuAlCr4_4S8_8 and (d) CuGaCr4_4S8_8. For evaluating IobsI_{\mathrm{obs}}, Lorentz factor correction is performed by multiplying sin⁡θsin⁡2θ\sin\theta\sin2\theta to the observed intensity. IcalI_{\mathrm{cal}} is calculated using the structural parameters at 4 K (Tables II and III). (e) Orientation of the incommensurate (qa′=qICq_{a'} = q_{\mathrm{IC}}) and commensurate (qb′=0.5q_{b'} = 0.5) magnetic modulation imposed on the orthorhombic cell.

RB≡∑K∣Iobs(K)−Ical(K)∣∑KIobs(K),R_{\mathrm{B}} \equiv\frac{\sum_K \lvert I_{\mathrm{obs}}(K)-I_{\mathrm{cal}}(K)\rvert}{\sum_K I_{\mathrm{obs}}(K)},

where the sums are taken over 8 nuclear reflections, is 2.78% and 4.14% for MM = Al and Ga, respectively. The obtained scale factors are used for the subsequent magnetic-structure analysis.

As shown in Figure 4(a) and Figure 4(b), we observe a number of additional Bragg reflections of magnetic origin at 2 K. All of theses peaks can be indexed by a single magnetic propagation vector, Q≈(0.89,0.11,0)c\mathbf{Q} \approx(0.89, 0.11, 0)_{\mathrm{c}} for CuAlCr4_4S8_8 and Q≈(0.81,0.19,0)c\mathbf{Q} \approx(0.81, 0.19, 0)_{\mathrm{c}} for CuGaCr4_4S8_8 in the cubic setting. The latter one is almost consistent with previously reported Q=(0.80,0.18,0)c\mathbf{Q} = (0.80, 0.18, 0)_{\mathrm{c}} for CuGaCr4_4S8_8.⁶ Remarkably, the identified Q\mathbf{Q} vectors can be represented as Q=(qIC,0.5,0)o\mathbf{Q} = (q_{\mathrm{IC}}, 0.5, 0)_{\mathrm{o}} in the orthorhombic setting, where qIC≈0.39q_{\mathrm{IC}} \approx0.39 for CuAlCr4_4S8_8 and qIC≈0.31q_{\mathrm{IC}} \approx0.31 for CuGaCr4_4S8_8. Considering the exchange-striction mechanism where the AFM configuration favors a shorter Cr–Cr bond length due to the predominant direct exchange coupling, we presume that the longer-period incommensurate modulation (qa′=qICq_{a'} = q_{\mathrm{IC}}) propagates along the longer diagonal (a′a' axis) and the shorter-period commensurate modulation (qb′=0.5q_{b'} = 0.5) propagates along the shorter diagonal (b′b' axis), as illustrated in Figure 4(e).

For the magnetic-structure analysis, we employ a least-squares refinement of the integrated intensities IobsI_{\mathrm{obs}} of magnetic Bragg peaks, as in Ref. 6. For MM = Al (Ga), we take into account 9 (13) observed peaks and 3 (1) unobserved peaks below 2θ=42∘2\theta= 42^\circ (44∘44^\circ), where the magnetic reflections are well separated with a few exceptions. In cases where multiple reflections at close 2θ2\theta values appear as a single peak in the ND profile, we take a sum of their calculated intensities (IcalI_{\rm cal}) and compare it with IobsI_{\rm obs}.

We calculate IcalI_{\rm cal} based on the orthorhombic crystal structure at 4 K revealed by the XRD (Tables II and III). In the following, we define 8 distinct spins within the orthorhombic unit cell as spins 1 ∼ 8, as depicted in Figure 5(a), where the breathing pyrochlore network is projected on the a′b′a'b' plane. We assume a coplanar structure with a single spiral plane expressed as

Definition of spins and magnetic structures

Figure 5. (a) Definition of spins 1 ∼ 8 in a breathing pyrochlore network projected on the a′b′a'b' plane. Odd (even) number spins belong to the Cr1 (Cr2) site [see Fig. 3(b)]. Blue and red squares represent small and large tetrahedra, respectively. Black dashed line represents an orthorhombic unit cell. (b) Schematic of an elliptical spiral plane for the tilted cycloidal structure (Model #4). mxm_x and mym_y represent magnetic moments parallel and perpendicular to the a′a' axis, respectively. θ\theta is an tilting angle from the a′b′a'b' plane. Model #2 is a special case of θ=90∘\theta= 90^\circ. [(c)(d)] IobsI_{\rm obs} versus IcalI_{\rm cal} plot of the magnetic Bragg peaks assuming Models #2 for (c) CuAlCr4_4S8_8 and (d) CuGaCr4_4S8_8. [(e)(f)] Schematics of the a′ca'c cycloidal structure obtained by a least-squares refinement for (e) CuAlCr4_4S8_8 and (f) CuGaCr4_4S8_8.

m(rl,d)=mxcos⁡(Q⋅rl,d+ϕd)+mysin⁡(Q⋅rl,d+ϕd),(1)\mathbf{m}(\mathbf{r}_{l,d}) = \mathbf{m}_x \cos(\mathbf{Q} \cdot\mathbf{r}_{l,d} + \phi_d) + \mathbf{m}_y \sin(\mathbf{Q} \cdot\mathbf{r}_{l,d} + \phi_d), \tag*{(1)}

where rl,d\mathbf{r}_{l,d} and ϕd\phi_d are the position vector and the phase, respectively, for the spin dd (d=1∼8d = 1 \sim8). Given that two tetrahedra, consist of spins 1 ∼ 4 and 5 ∼ 8, respectively, are of the same size, we add a constraint δn,4=δn+4,8\delta_{n,4} = \delta_{n+4,8} (n=1,2,3)n = 1, 2, 3), where δd,d′≡ϕd−ϕd′\delta_{d,d'} \equiv\phi_d - \phi_{d'} is defined as the phase difference between spins dd and d′d'. Here, we consider the case of δn,n+4=0\delta_{n,n+4} = 0 (n=1,2,3,4n = 1, 2, 3, 4), resulting in three independent parameters; δ1,4\delta_{1,4}, δ2,4\delta_{2,4}, and δ3,4\delta_{3,4}. Note that another constraint, δn,n+4=π\delta_{n,n+4} = \pi, can describe the identical magnetic structure obtained for δn,n+4=0\delta_{n,n+4} = 0 because of qb′=0.5q_{b'} = 0.5.

CuAlCr4_4S8_8CuGaCr4_4S8_8
Model #1 (a′b′a'b' cycloid)11.60%14.72%
Model #2 (a′ca'c cycloid)7.20%12.47%
Model #3 (proper screw)28.83%30.35%
Model #4 (tilted cycloid)6.92%12.18%

Table 4. Intensity reliability factors RBR_{\rm B} of a least-squares fit on the integrated intensity IobsI_{\rm obs} of magnetic Bragg peaks for CuMCr4_4S8_8 (MM = Al, Ga), assuming four types of elliptical helical magnetic structures (see the text for details).

Modelδ1,4\delta_{1,4}δ2,4\delta_{2,4}δ3,4\delta_{3,4}mxm_xmym_yθ\theta
MM = Al, #274(2)∘74(2)^\circ53(2)∘53(2)^\circ350(2)∘350(2)^\circ2.79(2) μB2.79(2)\ \mu_{\rm B}1.39(4) μB1.39(4)\ \mu_{\rm B}90∘90^\circ
MM = Al, #479(2)∘79(2)^\circ53(2)∘53(2)^\circ356(2)∘356(2)^\circ2.73(2) μB2.73(2)\ \mu_{\rm B}1.53(5) μB1.53(5)\ \mu_{\rm B}65(7)∘65(7)^\circ
MM = Ga, #240(2)∘40(2)^\circ10(2)∘10(2)^\circ330(2)∘330(2)^\circ3.09(2) μB3.09(2)\ \mu_{\rm B}1.63(5) μB1.63(5)\ \mu_{\rm B}90∘90^\circ
MM = Ga, #447(2)∘47(2)^\circ10(3)∘10(3)^\circ338(2)∘338(2)^\circ3.02(2) μB3.02(2)\ \mu_{\rm B}1.82(4) μB1.82(4)\ \mu_{\rm B}64(5)∘64(5)^\circ

Table 5. Magnetic structure parameters obtained by a least-squares refinement assuming the a′ca'c cycloidal (Model #2) and tilted cycloidal (Model #4) structures for CuMCr4_4S8_8 (MM = Al, Ga). For the definition of each parameter, see the text and Figure 5(b).

We consider three types of high-symmetry spiral planes, parallel to the a′b′a'b' plane (Model #1), the a′ca'c plane (Model #2), and the b′cb'c plane (Model #3). Since the incommensurate magnetic modulation propagates along the a′a' axis, Models #1 and #2 correspond to cycloidal structures, while Model #3 corresponds to the proper-screw type. An ellipticity is introduced so as to match the lattice Imm2Imm2 symmetry for each model, leading to two degrees of freedom regarding the moment size, mxm_x and mym_y, common to all the spins. Table IV summarizes RBR_{\rm B} values of the final solution obtained by refining five parameters, namely δ1,4\delta_{1,4}, δ2,4\delta_{2,4}, δ3,4\delta_{3,4}, mxm_x, and mym_y, for each model. For both the compounds, Model #2 yields the most satisfactory fit, as shown in the IobsI_{\rm obs} versus IcalI_{\rm cal} plots in Figure 5(c) and Figure 5(d). On the other hand, Model #3 yields a poor fit, suggesting that the ground state is a cycloidal structure but not a proper-screw type one.

Furthermore, we consider a tilting angle θ\theta of the spiral plane from the a′b′a'b' plane toward the a′ca'c plane, as shown in Figure 5(b). We call this “tilted cycloidal state” (Model #4), which interpolates Model #1 (θ=0∘\theta= 0^\circ) and Model #2 (θ=90∘\theta= 90^\circ). As shown in Table IV, the additional introduction of θ\theta results in a smaller RBR_{\rm B}, although the reduction of RBR_{\rm B} from Model #2 is less than 0.3%. Table V summarizes six parameter sets of the final solution assuming Models #2 and #4 for each compound. Here, mxm_x (mym_y) is the moment size parallel (perpendicular) to the a′a' axis. δ1,4∼δ3,4\delta_{1,4} \sim\delta_{3,4} converge to almost the same values regardless of the model. We hence conclude that the a′ca'c cycloidal state is the most probable structure. It is necessary to verify whether the magnetic moment along the b′b' component exists in the future study on single crystals. Figure 5(e) and Figure 5(f) schematically illustrate the magnetic structure of the a′ca'c cycloidal state for CuAlCr4_4S8_8 and CuGaCr4_4S8_8, respectively. The spiral plane is found to be elliptically elongated along the a′a' axis and contracted along the cc axis for both the compounds, i.e., mx>mym_x > m_y.

Discussion

We have revealed that CuMCr4_4S8_8 (MM = Al, Ga) undergoes a magnetic transition to cycloidal order accompanied by rhombic lattice distortion at low temperatures, where an incommensurate magnetic modulation propagates along the elongated ⟨110⟩\langle110\rangle axis of the original cubic cell. Let us now delve into the detailed discussion of the magnetic structure Figure 5(e) and Figure 5(f). In Figure 6(a) and Figure 6(b), we present all the relative angles between NN spins for M=AlM = \mathrm{Al} and Ga\mathrm{Ga}, respectively. The corresponding Cr–Cr bond lengths at 4 K, revealed by the crystal structure analysis assuming the Imm2Imm2 space group Figure 3(g) and Figure 3(k), are shown in Figure 6(c) and Figure 6(d) for M=AlM = \mathrm{Al} and Ga\mathrm{Ga}, respectively.

Figure 6. [(a)(b)] Relative angles of NN spins for (a) CuAlCr₄S₈ and (b) CuGaCr₄S₈. Blue and red tetrahedra correspond to small and large tetrahedra, respectively. [(c)(d)] NN Cr–Cr bond lengths at 4 K revealed by the powder XRD experiments for (c) CuAlCr₄S₈ and (d) CuGaCr₄S₈. Bonds of the same length are drawn in the same color (purple) and thickness.

First, it is evident that the spin configuration in the small tetrahedron is different from that in the large tetrahedron for both the compounds, reflecting the breathing nature of the pyrochlore lattice. For CuGaCr₄S₈, the averaged value of Si⋅Sj\boldsymbol{S}_i \cdot\boldsymbol{S}_j for all the NN spin pairs, ⟨Si⋅Sj⟩\langle\boldsymbol{S}_i \cdot\boldsymbol{S}_j\rangle, is 0.10 in the small tetrahedron and 0.75 in the large tetrahedron, assuming ∣Si∣=1|\boldsymbol{S}_i| = 1 (Note that −1/3≤⟨Si⋅Sj⟩≤1-1/3 \le\langle\boldsymbol{S}_i \cdot\boldsymbol{S}_j\rangle\le1 for a tetrahedron). DFT calculations predict J=9.8J = 9.8 K (AFM) and J′=−11.4J^{\prime} = -11.4 K (FM) at room temperature.[14] Considering that the volume ratio of the larger tetrahedron to the smaller one changes little across TNT_\mathrm{N} Figure 3(k), the predominantly FM character of the spin configuration in the large tetrahedron should be attributed to the strong FM coupling J′J^{\prime} below TNT_\mathrm{N}. For CuAlCr₄S₈, ⟨Si⋅Sj⟩\langle\boldsymbol{S}_i \cdot\boldsymbol{S}_j\rangle is 0.04 and 0.39 in the small and large tetrahedron, respectively, exhibiting a smaller difference compared to CuGaCr₄S₈. This may suggest that J′J^{\prime} in CuAlCr₄S₈ is either more weakly FM or, conversely, AFM. It is noteworthy that the breathing bond alternation is suppressed within the abab plane in the low-temperature phase in CuAlCr₄S₈, as mentioned in § 2.3. Consequently, the strengths of the exchange interactions may also change significantly across TNT_\mathrm{N}.

Next, we focus on the relation of the local spin configuration to the local lattice distortion. For both the compounds, a pair of spins on the dad_a bond (parallel to a′a^{\prime} axis) align rather ferromagnetically, while a pair on the dbd_b bond (parallel to the b′b^{\prime} axis) antiferromagnetically within the small tetrahedron Figure 6(a) and Figure 6(b). In contrast, a pair on the da′d^{\prime}_a bond (parallel to a′a^{\prime} axis) align rather antiferromagnetically, and a pair on the db′d^{\prime}_b bond (parallel to the b′b^{\prime} axis) ferromagnetically within the large tetrahedron. Notably, the FM bond is always longer than the AFM bond within the a′b′a^{\prime}b^{\prime} plane in each tetrahedron, i.e., da>dbd_a > d_b and da′<db′d^{\prime}_a < d^{\prime}_b Figure 6(c) and Figure 6(d). This trend is consistent with the exchange-striction picture for the Cr spinels, in which the AFM configuration favors a shorter Cr–Cr bond length. As for the remaining four bonds within each tetrahedron, which should be of the same length under the Imm2Imm2 symmetry, the relative angles between NN spins are not identical to each other. This is due to the broken mirror symmetry in the magnetic propagation vector Q=(qIC,0.5,0)o\boldsymbol{Q} = (q_\mathrm{IC}, 0.5, 0)_\mathrm{o}. We infer that the crystal structure below TNT_\mathrm{N} may indeed exhibit monoclinic I2I2 symmetry, although the current Rietveld analysis based on powder XRD does not allow the verification of additional crystallographic symmetry breaking.

Our magnetic-structure analysis also reveals that the averaged moment, m=[(mx)2+(my)2]/2m = \sqrt{[(m_x)^2 + (m_y)^2]/2}, is reduced from the full moment of 3 μB\mu_\mathrm{B} per Cr³⁺ spin: m≈2.20(3) μBm \approx2.20(3)\ \mu_\mathrm{B} for CuAlCr₄S₈ and m≈2.47(3) μBm \approx2.47(3)\ \mu_\mathrm{B} for CuGaCr₄S₈ Table V. Since the orbital degree of freedom is almost quenched in Cr³⁺, the reduction of the ordered moment would be attributed to quantum fluctuation. Nevertheless, we note that the ordered moment in CuMCr₄S₈ is larger than those in most Cr spinel oxides.[17, 18, 22] This would be because the introduction of the FM exchange interaction J′J^{\prime} instead of the AFM JJ in the large tetrahedron partially alleviates the magnetic frustration.

From a symmetry point of view, CuMCr₄S₈ is expected to exhibit cross-correlation responses among spin, charge, and lattice. As the temperature decreases, the crystal structure undergoes a nonpolar-to-polar transition associated with a magnetic transition at TNT_\mathrm{N}. Consequently, the averaged zz coordinate of the Cr³⁺ (S2−S^2-) ions deviates from z=0.5z = 0.5 (Tables II and III), facilitating spontaneous electric polarization PcP_c along the cc axis. The shift of the average Cr position is estimated to be approximately 0.02 Å for both the compounds, so that the emergence of substantial electric polarization is expected. Moreover, helimagnetism could also contribute to the magnetoelectric effect below TNT_\mathrm{N}. The magnetic-structure analysis indicates the development of a cycloidal structure with the spiral plane parallel to the a′ca^{\prime}c plane, where an incommensurate magnetic modulation propagates along the a′a^{\prime} axis. The crystallographic polarity along the cc axis may result in the Dzyaloshinskii–Moriya interaction[29, 30] which stabilizes the a′ca^{\prime}c cycloidal state.[31, 32] As an inverse effect, this cycloidal structure can generate uniform electric polarization PcP_c along the cc axis due to the spin-current mechanism,[33] which has been verified in various compounds.[34, 35, 36, 37, 38] The short period of the magnetic modulation, likely arising from the magnetic frustration, should be advantageous for enhancing PcP_c. The combination of the exchange-striction and the spin-current mechanism may bring about magnetoelectric effects in CuMCr₄S₈. Indeed, anomalous dielectric responses have been observed in polycrystalline CuGaCr₄S₈ samples associated with magnetic transitions at low temperatures and with the application of a magnetic field.[14] Other intriguing phenomena such as the electric-field control of magnetic domains and the piezoelectric effect may also be possible, although their verification on single crystals remains a future endeavor.

Summary

In summary, we have investigated the crystal and magnetic structures of the breathing pyrochlore magnets CuMCr₄S₈ (M=AlM = \mathrm{Al}, Ga\mathrm{Ga}) at low temperatures through the powder XRD and ND experiments, unveiling the emergence of a qualitatively similar ground state. At the magnetic transition temperature, CuMCr4S8\mathrm{CuMCr}_{4}\mathrm{S}_{8} undergoes a structural transition from cubic F‾43m\overline{F}43m to orthorhombic Imm2Imm2 symmetry, characterized by polar space group. The magnetic structure manifests an incommensurate cycloidal state with a magnetic propagation vector of Q=(qIC,0.5,0)oQ=(q_{\mathrm{IC}},0.5,0)_{\mathrm{o}} in the orthorhombic setting, where qIC≈0.39q_{\mathrm{IC}}\approx0.39 for M=AlM=\mathrm{Al} and qIC≈0.31q_{\mathrm{IC}}\approx0.31 for M=GaM=\mathrm{Ga}. The larger deviation of qICq_{\mathrm{IC}} from 0.5 for CuGaCr4S8\mathrm{CuGaCr}_{4}\mathrm{S}_{8} aligns with the observed larger rhombic distortion in terms of the exchange-striction mechanism. We also confirm a strong correlation between the local spin configuration and the local lattice distortion, underscoring the significance of spin-lattice coupling in determining the ground state. Remarkably, the low-temperature phase of CuMCr4S8\mathrm{CuMCr}_{4}\mathrm{S}_{8} is free from a phase separation often seen in other chromium spinels,[21, 22, 23, 24] suggesting that the identified spin-lattice-coupled helical magnetic order is robust against minor perturbations in the CuMCr4S8\mathrm{CuMCr}_{4}\mathrm{S}_{8} system. We propose that CuMCr4S8\mathrm{CuMCr}_{4}\mathrm{S}_{8} serves as a promising platform for investigating magnetoelectric effects arising from the emergent crystallographic polarity as well as helimagnetism.

Acknowledgments

This work was financially supported by the JSPS KAK-ENHI Grants-In-Aid for Scientific Research (No. 23K13068). The authors thank A. Ikeda for generously allowing the use of optical sensing instrument (Hyperion si155, LUNA) for thermal expansion measurements. The authors thank K. Adachi and D. Hashizume for the support of the in-house XRD experiments at Materials Characterization Support Team, RIKEN CEMS. The powder ND experiments at JRR-3 were carried out along the proposal No. 23401 and partly supported by ISSP of the University of Tokyo. The authors thank S. Kitou for fruitful discussions.

References

  1. [14]14) M. Gen, H. Ishikawa, A. Miyake, T. Yajima, H. O. Jeschke, H. Sagayama, A. Ikeda, Y. H. Matsuda, K. Kindo, M. Tokunaga, Y. Kohama, T. Kurumaji, Y. Tokunaga, and T. Arima, Breathing pyrochlore magnet CuGaCr₄S₈: Magnetic, thermodynamic, and dielectric properties, Phys. Rev. Mater. 7, 104404 (2023).
  2. [17]17) S. Gao, K. Guratinder, U. Stuhr, J. S. White, M. Mansson, B. Roessli, T. Fennell, V. Tsurkan, A. Loidl, M. Ciomaga Hatnean and G. Balakrishnan, S. Raymond, L. Chapon, V. O. Garlea, A. T. Savici, A. Cerverlino, A. Bombardi, D. Chernyschov, Ch. Rüegg, J. T. Haraldsen, and O. Zaharko, Manifolds of magnetic ordered states and excitations in the almost Heisenberg pyrochlore antiferromagnet MgCr₂O₄, Phys. Rev. B 97, 134430 (2018).
  3. [18]18) S. Ji, S.-H. Lee, C. Broholm, T. Y. Koo, W. Ratcliff, S.-W. Cheong, and P. Zschack, Spin-Lattice Order in Frustrated ZnCr₂O₄, Phys. Rev. Lett. 103, 037201 (2009).
  4. [21]21) M. Matsuda, H. Ueda, A. Kikkawa, Y. Tanaka, K. Katsumata, Y. Narumi, T. Inami, Y. Ueda, and S.-H. Lee, Spin-lattice instability to a fractional magnetization state in the spinel HgCr₂O₄, Nat. Phys. 3, 397 (2007).
  5. [22]22) R. Saha, F. Fauth, M. Avdeev, P. Kayser, B. J. Kennedy, and A. Sundaresan, Magnetodielectric Effects in A-site cation-ordered chromate spinels LiMCr₄O₈ (M=Ga and In), Phys. Rev. B 94, 064420 (2016).
  6. [23]23) G. J. Nilsen, Y. Okamoto, T. Masuda, J. Rodriguez-Carvajal, H. Mutka, T. Hansen, and Z. Hiroi, Complex magnetostructural order in the frustrated spinel LiInCr₄O₈, Phys. Rev. B 91, 174435 (2015).
  7. [24]24) F. Yokaichiya, A. Krimmel, V. Tsurkan, I. Margiolaki, P. Thompson, H. N. Bordallo, A. Buchsteiner, N. Stüßer, D. N. Argyriou, and A. Loidl, Phys. Rev. B 79, 064423 (2009).
  8. [29]29) I. Dzyaloshinskii, A thermodynamic theory of “weak” ferromagnetism of antiferromagnetics. J. Phys. Chem. Solids 4, 241 (1958).
  9. [30]30) T. Moriya, New Mechanism of Anisotropic Superexchange Interaction. Phys. Rev. Lett. 4, 228 (1960).DOI
  10. [31]31) E. Ruff, S. Widmann, P. Lunkenheimer, V. Tsurkan, S. Bordács, I. Kézsmárki, and A. Loidl, Multiferroicity and skyrmions carrying electric polarization in GaV₄S₈, Sci. Adv. 1, e1500916 (2015).DOI
  11. [32]32) Y. Fujima, N. Abe, Y. Tokunaga, and T. Arima, Thermodynamically stable skyrmion lattice at low temperatures in a bulk crystal of lacunar spinel GaV₄Se₈, Phys. Rev. B 95, 180410(R) (2017).DOI
  12. [33]33) H. Katsura, N. Nagaosa, and A. V. Balatsky, Spin Current and Magnetoelectric Effect in Noncollinear Magnets, Phys. Rev. Lett. 95, 057205 (2005).pubmed.ncbi.nlm.nih.gov/16090916
  13. [34]34) G. Lawes, A. B. Harris, T. Kimura, N. Rogado, R. J. Cava, A. Aharony, O. Entin-Wohlman, T. Yildirim, M. Kenzelmann, C. Broholm, and A. P. Ramirez, Magnetically Driven Ferroelectric Order in Ni₃V₂O₈, Phys. Rev. Lett. 95, 087205 (2005).pubmed.ncbi.nlm.nih.gov/16196898
  14. [35]35) T. Arima, A. Tokunaga, T. Goto, H. Kimura, Y. Noda, and Y. Tokura, Collinear to Spiral Spin Transformation without Changing the Modulation Wavelength upon Ferroelectric Transition in Tb₁₋ₓDyₓMnO₃, Phys. Rev. Lett. 96, 097202 (2006).pubmed.ncbi.nlm.nih.gov/16606307
  15. [36]36) Y. Yamasaki, S. Miyasaki, Y. Kaneko, J.-P. He, T. Arima, and Y. Tokura, Magnetic Reversal of the Ferroelectric Polarization in a Multiferroic Spinel Oxide, Phys. Rev. Lett. 96, 207204 (2006).pubmed.ncbi.nlm.nih.gov/16803202
  16. [37]37) T. Kimura, J. C. Lashley, and A. P. Ramirez, Inversion-symmetry breaking in the noncollinear magnetic phase of the triangular-lattice antiferromagnet CuFeO₂, Phys. Rev. B 73, 220401(R) (2006).DOI
  17. [38]38) S. Park, Y. J. Choi, C. L. Zhang, and S.-W. Cheong, Ferroelectricity in an S = 1/2 Chain Cuprate, Phys. Rev. Lett. 98, 057601 (2007).pubmed.ncbi.nlm.nih.gov/17358898

Paper details

vlnc:W_ykzzchnz
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents