betaFeedback

WorksPhysics and Astronomy

High-temperature magnetic anomaly via suppression of antisite disorder through synthesis route modification in a Kitaev candidate Cu2_2IrO3_3

Abstract

By incorporating inert KCl into the Na2_2IrO3_3 + 2CuCl →\to Cu2_2IrO3_3 + 2NaCl topochemical reaction, we significantly reduced the synthesis temperature of Cu2_2IrO3_3 from the 350∘^\circC reported in previous studies to 170∘^\circC. This adjustment decreased the Cu/Ir antisite disorder concentration in Cu2_2IrO3_3 from ∼\sim19%\% to ∼\sim5%\%. Furthermore, magnetic susceptibility measurements of the present Cu2_2IrO3_3 sample revealed a weak ferromagnetic-like anomaly with hysteresis at a magnetic transition temperature of ∼\sim70 K. Our research indicates that the spin-disordered ground state reported in chemically disordered Cu2_2IrO3_3 is an extrinsic phenomenon, rather than an intrinsic one, underscoring the pivotal role of synthetic chemistry in understanding the application of Kitaev model to realistic materials.

Introduction

In quantum spin liquids (QSL), strong quantum fluctuations suppress magnetic ordering, resulting in a quantum state in which the spins are strongly entangled [1, 2, 3]. As a result, novel quasiparticles emerge that are different from conventional magnons in their magnetic excitations [4, 5, 6]. A spin model recognized by Ising-like bond-dependent interactions on a tricoordinated lattice (Kitaev model) is the most promising spin model that realizes a highly entangled QSL as a ground state [7]. The elementary excitations in Kitaev QSL are described by itinerant Majorana fermions and localized Z2\mathbb{Z}_2 fluxes emergent from the fractionalization of quantum spins [7]. Furthermore, the Kitaev model is materialized by the superexchange interactions between spin-orbital-coupled Jeff=1/2J_{\mathrm{eff}} = 1/2 electrons in 4d/5d4d/5d transition metal ions with a low-spin d5d^5 electron configuration through the Jackeli-Khaliullin mechanism [8]. In this research trend, signatures of Kitaev QSL have been captured even in realistic materials [9, 10, 11], and then the materials developed to realize the Kitaev QSL ground state have been vigorously promoted worldwide [12, 13, 14, 15]. One of the problems in realistic Kitaev materials is that the chemical disorders obscure the actual ground state [16, 17, 18, 19]. Chemical disorders such as antisite disorder and stacking defects create a spatially fluctuating charge environment around magnetic ions, resulting in randomized pseudodipolar interactions. This randomness leads to a disordered state that mimics disordered states seemingly indistinguishable from quantum spin liquids in the form of superposition domains of short-range order [20]. In α\alpha-Li2IrO3\mathrm{Li}_2\mathrm{IrO}_3, as an "old-school" Kitaev material, non-negligible antisite disorder on the honeycomb layer has been confirmed [21]. The disorder is controlled to some extent by the synthesis temperature. The magnetic order is observed as a cusp in the temperature dependence of the magnetic susceptibility at 15 K for the sample with minor antisite disorder. In contrast, the magnetic cusp is obscured for the sample with more significant antisite disorders.

Topochemical reactions have emerged as a promising strategy for creating novel Kitaev materials [10, 11, 14, 22, 23, 24, 25]. This method involves substituting the alkali metal ions in old-school Kitaev materials with a range of different isovalent/aliovalent cations, all under mild temperature conditions, which broadens the material search space for potential Kitaev physics. However, it simultaneously introduces a dilemma: while expanding the variety of materials, it also intensifies significant chemical disorders, including antisite disorder [14]. Especially in Cu2IrO3\mathrm{Cu}_2\mathrm{IrO}_3, synthesized via an ion exchange reaction of Na2IrO3\mathrm{Na}_2\mathrm{IrO}_3 and CuCl at 350°C for 16 h, crystal structure analysis has clarified the presence of ~19% Cu and Ir antisite disorder [17, 23]. Furthermore, observing a spin-glass-like anomaly at 2.7 K also suggests the antisite disorder in Cu2IrO3\mathrm{Cu}_2\mathrm{IrO}_3. In addition, partial charge disproportionation has been observed as a chemical disorder specific to Cu2IrO3\mathrm{Cu}_2\mathrm{IrO}_3: about 10% of Cu+\mathrm{Cu}^{+} is oxidized to Cu2+\mathrm{Cu}^{2+}, and some Ir4+\mathrm{Ir}^{4+} is reduced to Ir3+\mathrm{Ir}^{3+} as charge compensation, as revealed by X-ray absorption spectroscopy experiments [26]. These facts demonstrate that the reaction temperature of 350°C is sufficient to produce these chemical disorders.

Meanwhile, attempts to observe the signatures of Kitaev QSL in Cu2IrO3\mathrm{Cu_2IrO_3} continue, even considering the effects of the chemical disorder [27, 28]. However, the "smoking gun" for the Kitaev QSL feasibility in Cu2IrO3\mathrm{Cu_2IrO_3} has not yet been found in the present stage. Therefore, a different approach will be needed to solve these problems. Under this circumstance, we attempted to suppress the chemical disorder by seeking gentler synthetic conditions.

Our study aimed to reduce the reaction temperature by employing CuCl-KCl eutectic salts as the ion exchanger for the topochemical reaction instead of using CuCl alone. The pseudobinary phase diagram of CuCl and KCl, as reconstructed from the previous report [29], is depicted in Figure 1. The melting points of CuCl and KCl stand at 426 ∘C426\,^{\circ}\mathrm{C} and 770 ∘C770\,^{\circ}\mathrm{C}, respectively, while their eutectic mixture melts at approximately 150 ∘C150\,^{\circ}\mathrm{C} [29]. Through this approach, we successfully decreased the synthesis temperature of Cu2IrO3\mathrm{Cu_2IrO_3} from 350 ∘C350\,^{\circ}\mathrm{C} to 170 ∘C170\,^{\circ}\mathrm{C}. This alternative approach significantly decreased the Cu/Ir antisite disorder within the honeycomb layer, reducing it from ∼19%\sim19\% to ∼5%\sim5\%. Moreover, magnetic susceptibility measurements of the resulting sample indicated a weak ferromagnetic-like anomaly around 7070 K, which contrasts with the low-temperature spin-glass behavior observed at 44 K in previously reported Cu2IrO3\mathrm{Cu_2IrO_3} samples. Therefore, our findings highlight the critical importance of addressing chemical disorders in Kitaev materials, as neglecting this issue can lead to incorrect assumptions regarding the ground state predicted by the Kitaev model.

Phase diagram of CuCl-KCl system

Figure 1. Phase diagram of CuCl-KCl system reconstructed from the previous report [29]. The gray and red stars indicate the temperature-composition synthesis conditions used in the topochemical reaction from Na2IrO3\mathrm{Na_2IrO_3} to Cu2IrO3\mathrm{Cu_2IrO_3} in previous reports and in our work, respectively.

Experimental Methods

A precursor material, Na2IrO3\mathrm{Na_2IrO_3}, was prepared using the conventional solid-state reaction method following the previous report [30]: Na2CO3\mathrm{Na_2CO_3} and Ir in the ratio of 1.05:11.05:1 were intimately mixed, and the pelletized mixture placed on the corundum crucible was heated at 825 ∘C825\,^{\circ}\mathrm{C} for 4848 h in air.

The initially reported synthesis of Cu2IrO3\mathrm{Cu_2IrO_3} was based on the topochemical reaction between precursor Na2IrO3\mathrm{Na_2IrO_3} and CuCl.

This study explored more mild synthetic conditions using CuCl-KCl eutectic salts instead of CuCl. Using mixed salts is expected to increase the ionic diffusion coefficient and promote ion exchange reactions at lower temperatures. The phase diagram of the CuCl-KCl system [29] shows that the melting point drops to ∼150 ∘C\sim150\,^{\circ}\mathrm{C} at a eutectic point of CuCl/(CuCl+KCl) ∼2/3\sim2/3 [see Fig. 1]. This precursor was ground well with a significant excess of CuCl-KCl mixture salt with the eutectic point ratio in an Ar-filled glove box, sealed in an evacuated Pyrex ampoule, and reacted at 170 ∘C170\,^{\circ}\mathrm{C} for 100100 h. In addition, the ion exchange reaction is hardly promoted at temperatures just above and below the nominal eutectic point (150±5 ∘C150\pm5\,^{\circ}\mathrm{C}). Thus, the reaction temperature is maintained slightly above the nominal eutectic point to expedite the reaction through complete melting. The ion-exchange reaction is expressed as

Na2IrO3+2CuCl→Cu2IrO3+2NaCl.(1)\mathrm{Na_2IrO_3 + 2CuCl \rightarrow Cu_2IrO_3 + 2NaCl}. \tag*{(1)}

After the reaction, the mixture was completely melted in the bottom of the Pyrex ampoule. The unreacted CuCl, inert KCl, and the byproduct NaCl were removed by washing the sample repeatedly with aqueous ammonia. The product was characterized by powder x-ray diffraction (XRD) experiments in a diffractometer with Cu-Kα\alpha radiation, and chemical analysis was conducted using a scanning electron microscope (JEOL IT-100) equipped with an energy dispersive x-ray spectroscope (EDX) with 1515 kV, 0.80.8 nA, 11-μ\mum beam diameter). The cell parameters and crystal structure were refined by the Rietveld method using the Z-RIETVELD software [31].

The temperature dependence of the magnetization was measured under several magnetic fields using the magnetic property measurement system (MPMS; Quantum Design) at the Institute for Solid State Physics at the University of Tokyo. Magnetization curves up to ∼60 T were measured by an induction method with a multilayer pulsed magnet at the International MegaGauss Science Laboratory of the Institute for Solid State Physics at the University of Tokyo. The temperature dependence of the heat capacity was measured using the conventional relaxation method in a physical property measurement system (PPMS; Quantum Design) at the Institute for Solid State Physics at the University of Tokyo.

Results and Discussion

The powder XRD pattern of Cu₂IrO₃ synthesized at 170°C is shown in Figure 2. Despite the lower synthesis temperature, no residual precursor and impurity have been observed. The characteristic asymmetric broadening of specific peaks observed within the 2θ2\theta range of 15–25°, as highlighted in the shaded area, and its enlarged view shown in Figure 2(b), is consistent with findings in previously reported Cu₂IrO₃ sample [23]. This broadening, also identified in similar layered materials, is attributed to stacking faults along the monoclinic cc-axis [32, 33] [34].

Powder X-ray diffraction patterns of Cu2IrO3

Figure 2. (a) The powder X-ray diffraction patterns of Cu2IrO3\mathrm{Cu_2IrO_3} synthesized at 170 ∘C170\,^{\circ}\mathrm{C}, showcasing observed intensities (in red), calculated intensities (in black), and their discrepancies (in blue). Vertical bars denote the Bragg reflection positions. The shaded region, where is excluded from the fitting range, emphasizes the characteristic asymmetric broadening of peak shapes, indicative of stacking faults. (b) The enlarged view of the asymmetric peak.

The crystal structure of Na₂IrO₃ was initially identified as C2/cC2/c [35], but subsequent studies reclassified it as C2/mC2/m [36, 37]. Given the similarity in the monoclinic structures of Na₂IrO₃ and Cu₂IrO₃, we conducted structural analyses for Cu₂IrO₃ employing both C2/cC2/c and C2/mC2/m models. These analyses yielded lower confidence parameters for the C2/mC2/m structure, with a reliable S-factor of ∼2.4 for the C2/cC2/c structure and ∼1.9 for the C2/mC2/m structure. The comprehensive results of the Rietveld analysis employing the C2/cC2/c model are extensively examined in the Supplemental Materials [38], which bolster the argument that the C2/mC2/m structural model rather than C2/cC2/c is more plausible from multiple chemical bonding viewpoints.

The chemical composition determined by EDX is 1.98(7) for Cu and 1.01(4) for Ir. Moreover, Na ion is not detected at all. These facts indicate that Cu₂IrO₃ is completely prepared even under gentle temperature conditions of 170°C. Our Rietveld refinement converges well with the crystal structure shown in Figure 3(a) and Figure 3(c) with crystallographic parameters in Table 1. In our sample, the extent of site mixing between Cu and Ir ions within the honeycomb layer has been significantly reduced to approximately 5%, marking a substantial improvement over a previously reported figure of about 19% [23]. This advancement indicates that synthesizing the material under gentler temperature conditions has effectively promoted the exchange of constituent ions while preserving the crystal structural integrity of materials with minimal changes. In our Cu₂IrO₃ sample, synthesized at a reaction temperature of 170°C, the three Ir-O-Ir bond angles are measured at 98.19° and 101.95°, and the two Ir-Ir bond distances are determined to be 3.14 Å and 3.09 Å, respectively. These variations in the Ir-O-Ir superexchange pathways will likely influence the strength of interactions within the material.

atomsiteggxxyyzzBB (Ų)
Ir14h4h0.9460(13)00.16843(9)1/21/20.117(12)
Ir22d2d0.108(3)00.51/21/20.40(4)
Cu14g4g100.3242(3)02.22(2)
Cu21a1a10000.11(3)
Cu32d2d0.892(3)00.51/21/20.40(4)
Cu44h4h0.0540(13)00.16843(9)1/21/20.117(12)
O14i4i10.1158(19)0.3268(5)0.248(8)
O28j8j10.1066(12)0.3228(4)0.3338(4)0.13(5)

Table 1. The crystal parameters of Cu₂IrO₃ (space group C2/mC2/m) determined from powder X-ray diffraction experiments with the monoclinic unit cell lattice parameters a=5.39170(14)a = 5.39170(14) Å, b=9.3369(2)b = 9.3369(2) Å, c=5.97308(7)c = 5.97308(7) Å and β=107.823(2)∘\beta= 107.823(2)^\circ. ‘g’ represents the atom occupancy parameter, and 'B' denotes the atomic displacement parameter.

Crystal structure and electron density distributions of Cu₂IrO₃

Figure 3. (a) Crystal structure of Cu₂IrO₃ (Trxn=170∘T_{\mathrm{rxn}} = 170^\circC) viewed parallel the honeycomb layer. The light-green, blue, and light-pink spheres represent Cu, Ir, and O ions. The parallelepiped drawn by solid lines indicates the unit cell. (b) Three-dimensional nuclear density distributions of Cu₂IrO₃ calculated by the maximum entropy method (MEM) through the Z-RIETVELD software [31]. (c) The honeycomb layer structure of Cu₂IrO₃.

The accuracy of the C2/mC2/m structural model is further substantiated by the maximum entropy method (MEM) analysis. As illustrated in Figure 3(b), the electron density contours effectively highlight the disparity between the electron densities associated with Ir and Cu atoms and the covalent characteristics of the Cu-O bond within the O-Cu-O dumbbell configuration. Additionally, it is noted in the Supplemental Materials that the O-Cu-O motifs are not observed in the MEM analysis outcomes when employing the C2/cC2/c structural model [38], underscoring the distinctiveness of the C2/mC2/m model in capturing these features.

In a comparative experiment, structural analysis of our synthesized samples, prepared using the same method previously reported, involving the reaction of Na₂IrO₃ and CuCl at 350°C for 16 hours, revealed approximately 19% antisite disorder between Ir and Cu sites within the honeycomb layer. This result is in good agreement with the previous report [23]. See the supplemental material for details on the structural analysis of our Trxn=350T_{\mathrm{rxn}} = 350°C sample [38]. These results indicated that the reaction temperature of 350°C is high-energy enough to generate some degree of Cu/Ir antisite disordering without thoroughly maintaining the Ir honeycomb framework. The bond valence sum (BVS) calculation for Ir ions from the refined structural parameters yields +4.09, consistent with the expected valence of +4. For comparison, the BVS calculation to the previously reported Cu₂IrO₃ sample (Trxn=350T_{\mathrm{rxn}} = 350°C) yields +3.72. This valence reduction is comparable to the results of X-ray absorption spectroscopy experiments [27].

Figure 4 shows the magnetic susceptibility χ\chi and its inversed 1/χ1/\chi data for Cu₂IrO₃ (Trxn=170T_{\mathrm{rxn}} = 170°C) measured at 7 T. A linear relationship in 1/χ1/\chi versus TT in the high-temperature region indicates the presence of the local magnetic moment. A Curie-Weiss (CW) fitting of the inverse susceptibility at 200–300 K yields an effective magnetic moment μeff=1.951(2) μB\mu_{\mathrm{eff}} = 1.951(2)\,\mu_{\mathrm{B}} and Weiss temperature θW=−108.8(6)\theta_{\mathrm{W}} = -108.8(6) K. The negative θW\theta_{\mathrm{W}} value indicates that the interaction between Ir⁴⁺ spins is predominantly antiferromagnetic. Furthermore, the μeff\mu_{\mathrm{eff}} value suggests a Jeff=1/2J_{\mathrm{eff}} = 1/2 pseudospin system with a Landé gg factor of g=2.25g = 2.25, which is enhanced by the spin-orbit interaction and is comparable to other iridates [35, 39, 40, 41], including Cu₂IrO₃ (Trxn=350T_{\mathrm{rxn}} = 350°C) [23]. The χ\chi data diverge from the Curie-Weiss (CW) fit at ∼150\sim150 K and exhibit a weak anomaly at Tmag∼70T_{\mathrm{mag}} \sim70 K, indicative of either short-range or weak antiferromagnetic order. This characteristic was not observed in previously reported Cu₂IrO₃ samples (Trxn=350T_{\mathrm{rxn}} = 350°C) [23]. The emergence of this feature is likely attributed to the qualitative reduction of antisite disorder in the Cu₂IrO₃ sample, suggesting that improved synthesis conditions can significantly affect the magnetic properties.

Temperature dependence of magnetic susceptibility and its inverse for Cu₂IrO₃

Figure 4. Temperature dependence of magnetic susceptibility χ\chi and its inverse 1/χ1/\chi measured at 7 T of Cu₂IrO₃ prepared at 170°C. The dashed lines indicate the result of Curie-Weiss fitting.

The observed short-range ordering behavior near TmagT_{\mathrm{mag}} in the

χ\chi data measured at 7 T closely resembles the short-range ordering identified in the promising Kitaev QSL candidate H₃LiIr₂O₆ [10]. To confirm the similarity in behavior, we conducted magnetization measurements on Cu₂IrO₃ (Trxn=170T_{\mathrm{rxn}} = 170°C) using various magnetic fields, the results of which are shown in Figure 5. Under a low magnetic field of 10 mT, the χ\chi curve increases at ∼\sim70 K with a thermal hysteresis between the zero-field-cooled (ZFC) and field-cooled (FC) data. With an increasing magnetic field, the increase of the FC χ\chi curve is suppressed and becomes almost indistinct at 7 T. A similar high-temperature magnetic anomaly was also observed in β\beta-Li₂IrO₃ when subjected to a low magnetic field at 100 K [42]. This phenomenon can be coherently attributed to the reorientation of the principal axes of magnetization, a mechanism closely associated with the emergence of Kitaev-like correlations [42]. Currently, single crystals of Cu₂IrO₃ are unavailable, precluding the possibility of conducting detailed experiments to investigate changes in the easy axis. However, given that Cu₂IrO₃ is classified as a Kitaev magnetic material akin to β\beta-Li₂IrO₃, it is plausible that similar magnetic anomalies could manifest in Cu₂IrO₃. The magnitude of a weak ferromagnetic moment does not differ for three differently prepared samples (see Supplemental Material [38]). On the other hand, as shown in the bottom panel, Cu₂IrO₃ (Trxn=350T_{\mathrm{rxn}} = 350°C) does not show such ferromagnetic enhancement, which is consistent with previous reports [23]. Hence, an emergent magnetic anomaly would be due to minimizing antisite disorder.

Temperature dependence of magnetic susceptibility in Cu₂IrO₃ prepared at 170°C

Figure 5. (Top panel) Temperature dependence of magnetic susceptibility χ\chi in Cu₂IrO₃ prepared at 170°C measured at several magnetic fields. Ferromagnetic-like anomaly observed at Tmag∼70T_{\mathrm{mag}} \sim70 K. (Bottom panel) The χ\chi data measured at 10 mT of Cu₂IrO₃ prepared at 350°C for comparison.

Figure 6 and its left inset show the isothermal magnetization MM curves at 100 and 4.2 K for Cu₂IrO₃ (Trxn=170T_{\mathrm{rxn}} = 170°C). The curve at 100 K is linear, while one at 4.2 K shows small spontaneous magnetization with hysteresis. This observation demonstrates the presence of a weak ferromagnetic moment accompanied by magnetic ordering. The spontaneous magnetic moment is about 6.8×10−4 μB6.8 \times10^{-4}\ \mu_{\mathrm{B}} per Ir⁴⁺ atoms. The origin of this tiny ferromagnetic moment will be discussed later. On the other hand, the absence of hysteresis in the MM data in Cu₂IrO₃ (Trxn=350T_{\mathrm{rxn}} = 350°C) at 4.2 K demonstrates that it is not a magnetically ordered state in contrast to Cu₂IrO₃ (Trxn=170T_{\mathrm{rxn}} = 170°C). The right inset of Figure 6 shows the isothermal magnetization curve up to 60 T measured by an induction method with a multilayer pulsed magnet. In this field region, no metamagnetic anomalies involving the breakdown of magnetic order commonly seen in Kitaev magnets are observed [43, 44].

Isothermal magnetization curves in Cu₂IrO₃

Figure 6. The isothermal magnetization MM curves in Cu₂IrO₃ (Trxn=170T_{\mathrm{rxn}} = 170°C) measured at 4.2 K (black) and 100 K (green) with the MM data in Cu₂IrO₃ prepared at 350°C for comparison. The upper inset shows the enlarged view of the magnetization curves near 0 T. The lower inset shows magnetization curve measured at 4.2 K under pulsed magnetic fields up to 60 T. The magnitudes are calibrated to the data measured under static fields up to 7 T.

Figure 7(a) presents the results of heat capacity measurements for Cu₂IrO₃ (Trxn=170T_{\mathrm{rxn}} = 170°C) and (Cu₃/₄Li₁/₄)₂SnO₃. In these measurements, stannate is used to estimate the phonon contribution of the iridate, based on its well-known success as an effective nonmagnetic analog of Kitaev candidate iridates such as α\alpha-Li₂IrO₃ [45], Na₂IrO₃ [45], and Ag₃LiIr₂O₆ [22]. The (Cu₃/₄Li₁/₄)₂SnO₃ powder samples used in these measurements were synthesized according to the method described in reference [46]. These measurements reveal that no distinct phase transition is detectable near TmagT_{\mathrm{mag}}. These observations resemble the magnetic anomalies observed at 100 K in β\beta-Li₂IrO₃ [42]. Similarly, phenomena like those observed in spin glasses, marked by ferromagnetic moments that elude detection in heat capacity measurements, have been reported for pyrochlore iridates in studies [47, 48]. Given that minor magnetic anomalies in pyrochlore iridates have been detected in heat capacity measurements using highly crystalline samples [49], it is plausible that magnetic anomalies at TmagT_{\mathrm{mag}} of Cu₂IrO₃ could also become observable with enhanced crystallinity and reduced stacking faults. The inset of Figure 7 illustrates the C/TC/T data in the low-temperature region as a function of T2T^2, suggesting the presence of a finite γ\gamma-term behavior. This behavior indicates residual entropy, pointing to unresolved magnetic degrees of freedom within the system. Whether this observation stems from a Kitaev QSL state remains unclear in the present stage. In any case, clarifying the nature of the TmagT_{\mathrm{mag}} anomaly is essential before any potential investigation into Kitaev QSL can proceed.

Heat capacity and magnetic entropy of Cu₂IrO₃

Figure 7. (a) The temperature dependence of the heat capacity divided by the temperature C/TC/T of Cu₂IrO₃. The inset shows the C/TC/T data at low temperatures as a function of T2T^2. Data is compared with the nonmagnetic analog (Cu₃/₄Li₁/₄)₂SnO₃ [46] for estimation of the phonon contribution Cphonon/TC_{\mathrm{phonon}}/T of Cu₂IrO₃. (b) The magnetic heat capacity Cmag/TC_{\mathrm{mag}}/T after the subtraction of Cphonon/TC_{\mathrm{phonon}}/T and the magnetic entropy SmagS_{\mathrm{mag}} for Cu₂IrO₃. The horizontal dashed line indicates the value of Smag=Rln⁡2S_{\mathrm{mag}} = R\ln2, which is the total magnetic entropy derived from Jeff=1/2J_{\mathrm{eff}} = 1/2 pseudospins.

The magnetic contribution Cmag/TC_{\mathrm{mag}}/T was obtained by subtracting this phonon contribution from the experimental data. As shown in Figure 7(b), the Cmag/TC_{\mathrm{mag}}/T curve exhibits a broad peak around Tmag∼70T_{\mathrm{mag}} \sim70 K, confirming that the magnetic anomaly in the χ\chi-data is of bulk origin. Here, the accuracy of the estimation of the magnetic heat capacity of Cu₂IrO₃ should be carefully considered. Notably, the asymptotic value of SmagS_{\mathrm{mag}}, calculated by integrating the magnetic Cmag/TC_{\mathrm{mag}}/T with respect to TT, coincides with the expected total magnetic entropy for Jeff=1/2J_{\mathrm{eff}} = 1/2 pseudospin (Rln⁡2=5.76R\ln2 = 5.76 mol⁻¹K⁻¹) at high temperatures, as shown in Figure 7(b). Additionally, the increase in Cmag/TC_{\mathrm{mag}}/T around 150 K is attributed to the development of short-range order, marking the point where the observed χ\chi data and CW curve start to diverge. These observations demonstrate that (Cu3/4_{3/4}Li1/4_{1/4})2_2IrO3_3 serves as an effective non-magnetic analog for providing the lattice heat capacity contribution of Cu2_2IrO3_3, despite differences in phonon contributions due to the varying masses of Cu and Li, as well as Ir and Sn.

The TmagT_{\mathrm{mag}} anomaly with a weak ferromagnetic moment probably originates from freezing a minor fraction of the spin degrees of freedom. This interpretation is consistent with the relatively low magnitude of the freezing/ordering moment, ascribed to the spontaneous magnetization of approximately 0.6% of the total magnetic moment of the Ir4+^{4+} ion with Jeff=1/2J_{\mathrm{eff}} = 1/2. According to the magnetic analog of the Clausius-Clapeyron equation, the change in magnetic entropy ΔS\Delta S at the first-order phase transition is directly proportional to the increase in magnetization ΔM\Delta M associated with spontaneous magnetization. Consequently, the latent heat associated with the transition at TmagT_{\mathrm{mag}} is undetectably small. Therefore, it is reasonable to infer that anomalies at TmagT_{\mathrm{mag}} would not result in detectable changes in entropy. To further understand the high-temperature magnetic anomaly in Cu2_2IrO3_3, whether it stems from the Kitaev model or spin-orbit entangled Jeff=1/2J_{\mathrm{eff}} = 1/2 pseudospin physics, there is an urgent need to develop a method for preparing single-crystal samples. This development is under investigation, highlighting the importance of advancing sample preparation techniques to clarify the underlying mechanisms of these magnetic phenomena. Nevertheless, the TmagT_{\mathrm{mag}} anomaly could serve as a crucial piece of the puzzle in understanding and resolving aspects of Kitaev physics. From a wider perspective, optimized topochemical reactions under milder temperature conditions unveil the true physical properties previously obscured by chemical disorder. Therefore, strategies aimed at reducing the synthesis temperature could expose the genuine ground states of materials resembling spin liquids, where the impact of chemical disorder has traditionally been underestimated. Consequently, our findings suggest that overlooking chemical disorders could result in misleading conclusions.

Summary

Applying CuCl-KCl eutectic salts in the ion-exchange process from Na2_2IrO3_3 to Cu2_2IrO3_3 markedly lowered the reaction temperature from 350°C to 170°C. This adjustment in the synthesis approach led to the production of a Cu2_2IrO3_3 sample with significantly reduced antisite disorder between Ir and Cu from ∼19% to ∼5%, as verified through structural analysis. This sample demonstrated a magnetic anomaly near 70 K, featuring a slight ferromagnetic moment, marking a clear distinction from the spin glassy nonmagnetic ground state previously identified in Cu2_2IrO3_3 samples synthesized at 350°C. These findings underscore the critical role of synthesis conditions in defining the accurate structural and magnetic characteristics of complex oxides, suggesting that meticulous control over these parameters can unveil distinct physical properties.

Acknowledgment

This work was supported by JST PRESTO Grant Number JPMJPR23Q8 (Creation of Future Materials by Expanding Materials Exploration Space) and JSPS KAKENHI Grant Numbers. JP23H04616 (Transformative Research Areas (A) "Supra-ceramics"), JP22K14002 (Young Scientific Research), and JP21K03441 (Scientific Research (C)). Part of this work was carried out by joint research in the Institute for Solid State Physics, the University of Tokyo (Project Numbers 202012-HMBXX-0021, 202112-HMBXX-0023, 202106-MCBXG-0065 and 202205-MCBXG-0063).

References

  1. [1]L. Balents, Nature (London) 464, 199 (2010).
  2. [2]L. Savary and L. Balents, Rep. Prog. Phys. 80, 016502 (2016).
  3. [3]P. W. Anderson, Mater. Res. Bull. 8, 153 (1973).
  4. [4]T.-H. Han, J. S. Helton, S. Chu, D. G. Nocera, J. A. Rodriguez-Rivera, C. Broholm, and Y. S. Lee, Nature (London) 492, 406 (2012).
  5. [5]O. I. Motrunich, Phys. Rev. B 72, 045105 (2005).
  6. [6]S.-S. Lee and P. A. Lee, Phys. Rev. Lett. 95, 036403 (2005).
  7. [7]A. Kitaev, Ann. Phys. 321, 2 (2006).
  8. [8]G. Jackeli and G. Khaliullin, Phys. Rev. Lett. 102, 017205 (2009).
  9. [9]Y. Kasahara, T. Ohnishi, Y. Mizukami, O. Tanaka, S. Ma, K. Sugii, N. Kurita, H. Tanaka, J. Nasu, Y. Motome, T. Shibauchi, and Y. Matsuda, Nature (London) 559, 227 (2018).
  10. [10]K. Kitagawa, T. Takayama, Y. Matsumoto, A. Kato, R. Takano, Y. Kishimoto, S. Bette, R. Dinnebier, G. Jackeli, H. Takagi, Nature (London) 554, 341 (2018).
  11. [11]Y. Haraguchi, A. Matsuo, K. Kindo, and H. Aruga Katori, Phys. Rev. Materials 6, L021401 (2022).
  12. [12]H. Takagi, T. Takayama, G. Jackeli, G. Khaliullin, and S. E. Nagler, Nat. Rev. Phys. 1, 264 (2019).
  13. [13]Y. Motome and J. Nasu, J. Phys. Soc. Jpn. 89, 012002 (2020).
  14. [14]A. A. Tsirlin, P. Gegenwart, Phys. Status Solidi B 2021, 2100146 (2021).
  15. [15]F. Bahrami, M. Abrambchuk, O. Lebedev, and F. Tafti, Molecules 27, 871 (2022).
  16. [16]J. Nasu and Y. Motome, Phys. Rev. B 102, 054437 (2020).
  17. [17]Y. S. Choi, C. H. Lee, S. Lee, S. Yoon, W.-J. Lee, J. Park, A. Ali, Y. Singh, J.-C. Orain, G. Kim, J.-S. Rhy ee, W.-T. Chen, F. Chou, and K.-Y. Choi, Phys. Rev. Lett. 122, 167202 (2019).
  18. [18]S.-H. Do, C. H. Lee, T. Kihara, Y. S. Choi, S. Yoon, K. Kim, H. Cheong, W.-T. Chen, F. Chou, H. Nojiri, and K.-Y. Choi, Phys. Rev. Lett. 124, 047204 (2020).
  19. [19]W.-H. Kao, J. Knolle, G. B. Halász, R. Moessner, and N. B. Perkins, Phys. Rev. X 11, 011034 (2021).
  20. [20]Z. Zhu, P. A. Maksimov, S. R. White, and A. L. Chernyshev, Phys. Rev. Lett. 119, 157201 (2017).
  21. [21]H. Kobayashi, M. Tabuchi, M. Shikano, H. Kageyama, and R. Kanno, J. Mater. Chem. 13, 957 (2003).
  22. [22]F. Bahrami, W. Lafargue-Dit-Hauret, O. I. Lebedev, R. Movshovich, H.-Yu. Yang, D. Broido, X. Rocquefelte, and F. Tafti, Phys. Rev. Lett. 123, 237203 (2019).
  23. [23]M. Abramchuk, C. Ozsoy-Keskinbora, J.W. Krizan, K. R. Metz, D. C. Bell, and F. Tafti, J. Am. Chem. Soc. 139, 15371 (2017).
  24. [24]Y. Haraguchi, C. Michioka, A. Matsuo, K. Kindo, H. Ueda, and K. Yoshimura, Phys. Rev. Mater. 2, 054411 (2018).
  25. [25]Y. Haraguchi and H. Aruga Katori, Phys. Rev. Mater. 4, 044401 (2020).
  26. [26]E. M. Kenney, C. U. Segre, W. Lafargue-Dit-Hauret, O. I. Lebedev, M. Abramchuk, A. Berlie, S. P. Cottrell, G. Simutis, F. Bahrami, N. E. Mordvinova, G. Fabbris, J. L. McChesney, D. Haskel, X. Rocquefelte, M. J. Graf, and F. Tafti, Phys. Rev. B 100, 094418 (2019).
  27. [27]S. K. Takahashi, J. Wang, A. Arseneault, T. Imai, M. Abramchuk, F. Tafti, P. M. Singer, Phys. Rev. X, 9, 031047 (2019).
  28. [28]S. Pal, A. Seth, P. Sakrikar, A. Ali, S. Bhattacharjee, DVS Muthu, Y. Singh, A.K. Sood, Phys. Rev. B 104 184 (2021).
  29. [29]D.E. Etter, C.J. Wiedenheft, Sol. Energy Mater 2 423 (1980).
  30. [30]J. W. Krizan, J. H. Roudebush, G. M. Fox, and R. J. Cava, Mater. Res. Bull. 52, 162 (2014).
  31. [31]R. Oishi, M. Yonemura, Y. Nishimaki, S. Torii, A. Hoshikawa, T. Ishigaki, T. Morishima, K. Mori, and T. Kamiyama, Nucl. Instrum. Methods Phys. Res., Sect. A 600, 94 (2009).
  32. [32]B. E. Warren, Phys. Rev. 9, 693 (1941).
  33. [33]S. Bette, T. Takayama, V. Duppel, A. Poulain, H. Takagi, and R. E. Dinnebier, Dalton Trans. 48, 9250 (2019).
  34. [35]Y. Singh and P. Gegenwart, Phys. Rev. B 82, 064412 (2010).
  35. [36]F. Ye, S. Chi, H. Cao, B. C. Chakoumakos, J. A. Fernandez-Baca, R. Custelcean, T. F. Qi, O. B. Korneta, and G. Cao, Phys. Rev. B 85, 180403 (2012).
  36. [37]S. K. Choi, R. Coldea, A. N. Kolmogorov, T. Lancaster, I. I. Mazin, S. J. Blundell, P. G. Radaelli, Y. Singh, P. Gegenwart, K. R. Choi, S.-W. Cheong, P. J. Baker, C. Stock, and J. Taylor, Phys. Rev. Lett. 108, 127204 (2012).
  37. [38]See Supplemental Material at [URL will be inserted by publisher] for the sample dependence.
  38. [39]V. Todorova, A. Leineweber, L. Kienle, V. Duppel, and M. Jansen, J. Solid State Chem. 184, 1112 (2011).
  39. [40]Y. Okamoto, M. Nohara, H. Aruga Katori, and H. Takagi, Phys. Rev. Lett. 99, 137207 (2007).
  40. [41]N. Khan, D. Prishchenko, Y. Skourski, V. G. Mazurenko, and A. A. Tsirlin, Phys. Rev. B 99, 144425 (2019).
  41. [42]A. Ruiz, V. Nagarajan, M. Vranas, G. Lopez, G. T. McCandless, I. Kimchi, J. Y. Chan, N. P. Breznay, A. Franó, B. A. Frandsen, and J. G. Analytis, Phys. Rev. B 101, 075112 (2020).
  42. [43]Y. Kubota, H. Tanaka, T. Ono, Y. Narumi, and K. Kindo, Phys. Rev. B 91, 094422 (2015).
  43. [44]L. Janssen and M. Vojta, J. Phys.: Condens. Matter 31, 423002 (2019).
  44. [45]K. Mehlawat, A. Thamizhavel, and Y. Singh, Phys. Rev. B 95, 144406 (2017).
  45. [46]M. Abramchuk, O. I. Lebedev, O. Hellman, F. Bahrami, N. E. Mordvinova, J. W. Krizan, K. R. Metz, D. Broido, and F. Tafti, Inorg. Chem. 57, 12709 (2018).
  46. [47]N. Aito, M. Soda, Y. Kobayashi, and M. Sato, J. Phys. Soc. Jpn. 72, 1226 (2003).
  47. [48]N. Taira, M. Wakeshima, and Y. Hinatsu, J. Phys. Condens. Matter 13, 5527 (2001).
  48. [49]M. Klicpera, K. Vlášková, and M. Diviš, J. Magn. Magn. Mater. 506, 166793 (2020).

Paper details

vlnc:W_svetdgnc
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