betaFeedback

WorksPhysics and Astronomy

Magnetic properties of S = 5/2 anisotropic triangular chain Bi3FeMo2O12

Abstract

Competing magnetic interactions in low-dimensional quantum magnets can lead to the exotic ground state with fractionalized excitations. Herein, we present our results on an S = 5/2 quasi-one-dimensional spin system Bi3FeMo2O12. The structure of Bi3FeMo2O12 consists of very well separated, infinite zig-zag S = 5/2 spin chains. The observation of a broad maximum around 10 K in the magnetic susceptibility suggests the presence of short-range spin correlations. Magnetic susceptibility data do not fit to S=5/2 uniform spin chain model due to the presence of 2nd nearest-neighbor coupling (J2) along with the 1st nearest-neighbor coupling J1 of the zig-zag chain. The electronic structure calculations infer that the value of J1 is comparable with J2 (J2/J1~1.1) with a negligible inter-chain interaction (J'/J ~ 0.01), implying that Bi3FeMo2O12 is a highly frustrated triangular chain system. The absence of magnetic long-range ordering down to 0.2 K is seen in the heat capacity data, despite a relatively large antiferromagnetic Curie-Weiss temperature of -40 K. The magnetic heat capacity follows nearly a linear behavior at low temperatures indicating that the S = 5/2 anisotropic triangular chain exhibits the gapless excitations.

Introduction

Investigating the exotic magnetic properties of low-dimensional and geometrically frustrated spin systems is one of the active research fields in modern condensed matter physics [1, 2, 3, 4]. Mermin-Wagner theorem states that the system with dimensionality d≤2d \le2 and finite range interactions preserve continuous symmetry [5]. The quantum fluctuations, in general, originated from quantum effects, are intrinsic and significant in low-dimensional magnetic systems (LDMS). These are further prominent for low spin (S=1/2S = 1/2) magnetic materials. The physics of S=1/2S = 1/2 LDMS is quite rich, and they offer a viable ground for the experimental realization of correlated quantum states with exotic fractional excitations [1, 2]. The algebraic spin-spin correlation decay in S=1/2S = 1/2 uniform spin chain systems suggests that the ground state is gapless [6, 7, 8]. Further, the introduction of geometric frustration through the presence of 2nd2^{\mathrm{nd}} nearest neighbor (NN) interaction (J2J_2) along with that of 1st1^{\mathrm{st}} NN interaction (J1J_1) to the S=1/2S = 1/2 uniform spin chain leads to a gapped excitation spectrum in the ground state as per the exactly solvable Majumdar-Ghosh (MG) chain model [9]. The MG chain model states that J2/J1=0.5J_2/J_1 = 0.5 opens a spin-gap and forms a singlet ground state [10].

On the other hand, the LDMS with large spin (i.e., S=5/2S = 5/2) has not been studied extensively as the quantum effects are not prominent in these materials. A few examples with S=5/2S = 5/2 systems are SrMn₂V₂O₈ [11], SrMn(VO₄)(OH) [12], Ba₃Fe₂Ge₄O₁₄ [13], and Bi₂Fe(SeO₃)OCl₃ [14]. For example, the Mn-based linear chain system SrMn₂V₂O₈ exhibits a broad maximum (Tmax⁡T^{\max}) in the susceptibility data χ(T)\chi(T) around 200 K. However, due to the presence inter-chain couplings (J′/J1≥0.6J'/J_1 \ge0.6), this material shows a magnetic long-range order (LRO) at 45 K. The Fe-based zig-zag chain Bi₂Fe(SeO₃)OCl₃ system shows Tmax⁡T^{\max} around 130 K in the χ(T)\chi(T) data and LRO at 13 K, even in the presence of magnetic frustration with J2/J1≈0.2J_2/J_1 \approx0.2. All these S=5/2S = 5/2 spin systems undergo finite temperature due to the inter-chain coupling and insufficient magnetic frustration. It is pertinent to ask whether a correlated dynamic ground state is realizable or not in low dimensional systems with large spin. In this context, exploring novel low dimensional spin systems with large spin S=5/2S = 5/2 promising to host quantum spin liquid with exotic fractional excitations set an attractive setting.

In this paper, we report the magnetic properties and electronic structure calculations on Bi₃FeMo₂O₁₂ [15]. This material comprises the very well separated S=5/2S = 5/2 zig-zag chains passing along the cc-axis. The magnetic moments (S=5/2S = 5/2) interact antiferromagnetically with Curie-Weiss temperature θCW≈−40\theta_{\mathrm{CW}} \approx-40 K. No magnetic LRO or spin freezing is observed down to 0.2 K. From the electronic structure calculations, the estimated ratio of 2nd2^{nd} NN and 1st1^{st} NN couplings between Fe atoms is close to 1.1, and negligible inter-chain coupling (J′/J1≈0.01J^{\prime}/J_{1} \approx0.01), which suggests that the Bi3_{3}FeMo2_{2}O12_{12} is a unique material with very well separated S=5/2S = 5/2 triangular chains with a small anisotropy. Interestingly, heat capacity shows a nearly linear behavior with a finite value of the linear coefficient, suggesting the gapless excitations in the S=5/2S = 5/2 triangular chains, unlike the S=1/2S = 1/2 triangular chains that host a spin-gap ground state [16].

Experimental details

The polycrystalline samples of Bi3_{3}FeMo2_{2}O12_{12} and the non-magnetic analog Bi3_{3}GaMo2_{2}O12_{12} were synthesized by solid-state reaction method using the respective chemicals of Bi2_{2}O3_{3}, Fe2_{2}O3_{3}, Ga2_{2}O3_{3}, and MoO3_{3}. These chemicals are mixed in stoichiometric ratio and thoroughly grounded in agate mortar and pestle. The pellets were made using the hydraulic press and heated at different temperatures from 400°C to 800°C. Finally, the sample was fired at 800°C for 72 hours with a few intermediate grindings to obtain the single phase of the samples Bi3_{3}FeMo2_{2}O12_{12}. Neutron powder diffraction (NPD) experiments were performed using the neutron powder diffractometer PD-I (λ=1.094\lambda= 1.094 Å) with three linear position-sensitive detectors at Dhruva reactor, Bhabha Atomic Research Center, India. Magnetization (MM), and heat capacity (CpC_{p}) measurements were performed on the polycrystalline sample pellets using the Physical Properties Measurement System (PPMS) with the corresponding attachments of Vibration Sample Magnetometer (VSM) and heat-capacity measurement option, respectively, in the temperature range from 2 K to 300 K and in the magnetic fields up to 160 kOe. Low-temperature heat capacity data was measured using a dilution fridge on a flat pellet of Bi3_{3}FeMo2_{2}O12_{12}.

Results

Structural details

Bi3_{3}FeMo2_{2}O12_{12} crystallizes in a monoclinic structure with a space group C2/cC2/c and holds the Scheelite-type structure with ABO4_{4} family [15]. The Bi atoms are located at the A-site, while B-site is occupied by Fe and Mo atoms in Bi3_{3}FeMo2_{2}O12_{12} structure. Interestingly, the Fe and Mo atoms are ordered at the B-site. The unit cell consists of FeO4_{4}, MoO4_{4} tetrahedral, and BiO6_{6} polyhedral units (see Figure 1(a)). The obtained lattice parameters from the Rietveld refinement of XRD pattern are a=16.91a = 16.91 Å, b=11.65b = 11.65 Å, and c=5.25c = 5.25 Å, α=γ=90∘\alpha= \gamma= 90^\circ, β=107.1∘\beta= 107.1^\circ. The Fe3+^{3+} (S=5/2S = 5/2) ions form an infinite zig-zag chain running along the cc-axis, as shown in Figure 1(b). The 1st1^{st} NN distance of Fe-Fe is 3.79 Å with a possible exchange path of Fe-O1-O1-Fe. The 2nd2^{nd} NN distance of Fe-Fe is 5.25 Å, and its possible exchange path could be through Fe-O2-O2-Fe interactions. The bond lengths

Crystal structure and zig-zag chain of Bi3FeMo2O12

Figure 1. (Color online) (a) Unit cell of Bi3_{3}FeMo2_{2}O12_{12} [15]. (b) FeO4_{4} tetrahedral units form the zig-zag chain running along cc-axis. The 1st1^{st} NN Fe-Fe distance is 3.8 Å (with exchange coupling J1J_{1}) and 2nd2^{nd} NN Fe-Fe distance is 5.3 Å (with exchange coupling J2J_{2}).

and bond angles are shown in the table I. These chains are very well separated by a relatively large distance of 8.64 Å, suggesting that the compound might have nearly isolated S=5/2S = 5/2 zig-zag spin chains.

Neutron diffraction measurements

The data were recorded at different temperatures from 6 K to 50 K. Rietveld refinement of NPD data was performed using the FullProf Suite software package as shown in Figure 2(a), (Figure 2b), and (Figure 2c). The observed neutron diffraction pattern could be fitted by considering only the nuclear phase. We have subtracted the intensities of 50 K from 6 K data, i.e., II (50 K) −- II (6 K). We do not see any signs of new Bragg intensities or diffuse background patterns from the subtracted data as shown in Figure 2(d). We have also compared the difference plot of II (50 K) −- II (6 K) with the difference plot of II (50 K) −- II (20 K), and there is no difference seen between these two plots. Neither additional magnetic Bragg peaks nor an enhancement in the intensity of the fundamental nuclear Bragg peaks has been observed down to 6 K (see Figure 2(d)), indicating the absence of magnetic long-range order and rule out the presence of a phase transition at 11 K as was reported in reference [17].

Rietveld refinement of neutron diffraction data at different temperatures

Figure 2. (Color online) Rietveld refinement of Neutron diffraction (ND) data measured at different temperatures 50 K, 20 K, and 6 K are shown in (a), (b), and (c), respectively. Red circles represent the measured data (IobsI_{obs}), the black line represents the calculated diffraction pattern (IcalI_{cal}), the blue line represents the difference (Iobs−IcalI_{obs}-I_{cal}), and the green vertical lines represent the Bragg positions. (d) The ND intensities of 6 K and 20 K after subtracting the ND intensities at 50 K.

Magnetization measurements

Temperature-dependent magnetic susceptibility χ(T)\chi(T) measurements were performed on the polycrystalline sample Bi3_{3}FeMo2_{2}O12_{12} in the TT range from 2 - 300 K in H=10H = 10 kOe (see Figure 3(a)). The fit of the χ(T)\chi(T) data to Curie-Weiss law yields the temperature-independent susceptibility χ0≈−2.0×10−4\chi_{0} \approx-2.0 \times10^{-4} cm3^{3}/mol, Curie-Weiss temperature θCW≈−40\theta_{\mathrm{CW}} \approx-40 K, and Curie constant C≈4.3C \approx4.3 cm3^{3} K/mol. Diamagnetic susceptibility is estimated to be χdia≈2.4×10−4\chi_{\mathrm{dia}} \approx2.4 \times10^{-4} cm3^{3}/mol from the individual ions in the formula Bi3_{3}FeMo2_{2}O12_{12}. After the subtraction of χdia\chi_{dia} from the obtained value of χ0\chi_{0}, the calculated Van Vleck susceptibility is χvv≈4.2×10−5 cm3/mol\chi_{vv} \approx4.2 \times10^{-5}\,\mathrm{cm}^{3}/\mathrm{mol}. From the value of CC, the effective magnetic moment of Fe3+\mathrm{Fe}^{3+} is calculated to be 5.90 μB5.90\,\mu_{B} (=8C μB= \sqrt{8C}\,\mu_{B}), well consistent with the expected value of 5.91 μB5.91\,\mu_{B} for S=5/2S = 5/2. The absence of splitting in zero-field-cooled (ZFC) and field-cooled (FC) susceptibility rules out the spin-glass transition in this compound, as shown in the inset of Figure 3(a). Figure 3(b) represents the magnetization isotherm at 2 K was measured up to 160 kOe. The M(H)M(H) follows linear behavior, indicating the absence of ferromagnetic components in the samples. M(H)M(H) data do not saturate up to 160 kOe field. The large magnetic field is required to reach saturated magnetization Msat=gS=5μBM_{sat} = gS = 5\mu_{B} for S=5/2S = 5/2 magnetic moments.

Temperature-dependent magnetic susceptibility and magnetization isotherm

Figure 3. (Color online) (a) Temperature-dependent magnetic susceptibility from 2 K to 300 K. The inset of figure (a) shows the ZFC and FC magnetic susceptibility data under 100 Oe field. (b) Magnetic isotherm measured at T=2T = 2 K with variation of field 160 kOe. The inset shows the comparison of experimental data with S=5/2S = 5/2 uniform spin chain simulation for J/kB≈−1J/k_{B} \approx-1 K.

At low-TT, χ(T)\chi(T) shows a broad maximum around Tmax≈10T^{max} \approx10 K, indicating the presence of short-range correlations [6, 18, 19, 20]. We tried to analyze χ(T)\chi(T) with S=5/2S = 5/2 uniform spin chain model by Bonner and Fisher (BF) [21, 22] and found that this model could not reproduce our experimental data. According to the S=5/2S = 5/2 uniform spin chain, the broad maximum would appear at kBTmax/J=10.6k_{B}T^{max}/J = 10.6 [23]. From our experimental observation of TmaxT^{max} position, the J/kBJ/k_{B} value expected to be about −1-1 K [23]. We have then compared the experimental data with the S=5/2S = 5/2 uniform chain model with J/kB≈−1J/k_{B} \approx-1 K. As shown in the inset of Figure 3(b), the experimental value of magnetic susceptibility is much smaller than the simulated data, suggesting the presence of significant additional antiferromagnetic exchange couplings. From this analysis, we anticipated the presence of a significant value of 2nd2^{nd} NN coupling along with 1st1^{st} NN coupling qualitatively. This is further quantitatively confirmed from the first-principle density functional theory (DFT) electronic structure calculations discussed later in the paper. The presence of 2nd2^{nd} NN exchange coupling accounts for the large magnetic frustration in this material since J1J_{1}, and J2J_{2} form a triangular network (see Figure 1(b)).

Heat capacity measurements

The heat capacity Cp(T)C_{p}(T) data of Bi3_{3}FeMo2_{2}O12_{12} and Bi3_{3}GaMo2_{2}O12_{12} were investigated in zero field (see Figure 4). At low-TT, there is a large difference seen between the CpC_{p} of Bi3_{3}FeMo2_{2}O12_{12} and Bi3_{3}GaMo2_{2}O12_{12} indicating the dominance of magnetic contribution. Interestingly, no sharp peak is observed down to 0.2 K in the Cp(T)C_{p}(T) data of Bi3_{3}FeMo2_{2}O12_{12}, suggesting the absence of magnetic LRO. The frustration parameter f=∣θCW∣/TNf = \lvert\theta_{CW} \rvert/ T_{N} value is greater than 200, indicating the presence of strong spin frustration. To extract the magnetic contribution Cm(T)C_m(T), we have used the CpC_p data of its non-magnetic analog Bi3_3GaMo2_2O12_{12} [24]. CpC_p data of Bi3_3GaMo2_2O12_{12} is fitted with Debye expression [25] below 5 K.

Temperature-dependent heat capacity data of Bi3FeMo2O12 and Bi3GaMo2O12 with lattice part extracted from Debye fit

Figure 4. (Color online) Temperature-dependent heat capacity Cp(T)C_p(T) data of Bi3_3FeMo2_2O12_{12} and Bi3_3GaMo2_2O12_{12} with lattice part of the heat capacity data (red) extracted from Debye fit. The observed lattice contribution of the non-magnetic analog is extremely small at low temperatures, compared to the heat capacity of Bi3_3FeMo2_2O12_{12}. Inset shows the magnetic heat capacity CmC_m (left) and normalized magnetic entropy ΔSm\Delta S_m (right) versus TT.

Cph(T)=9R∑cn(TθDn)3∫x4ex(ex−1)2 dxC_{\mathrm{ph}}(T)=9R\sum c_n\left(\frac{T}{\theta_{D_n}}\right)^3\int\frac{x^4e^x}{(e^x-1)^2}\,dx

Here, the θDn\theta_{D_n} represent the Debye temperatures, cnc_n indicate the multiplication coefficients, and RR is the universal gas constant. The extracted Debye temperatures are θD1≈240\theta_{D1}\approx240 K and θD2≈308\theta_{D2}\approx308 K. The fit is extrapolated down to 0.2 K, and then subtracted from the CpC_p data of Bi3_3FeMo2_2O12_{12}. As shown in the inset of Figure 4, the Cm(T)C_m(T) data show a broad maximum at 6 K, like at 10 K in the χ(T)\chi(T) data. The appearance of a broad maximum in the CpC_p data is due to the short-range spin correlations that arise from the one-dimensional nature of exchange interaction between Fe3+^{3+} moments [6, 18]. It has been noticed in many LDMS that the broad maximum in the heat capacity is generally at low temperatures than that of the broad maximum in susceptibility [6, 26]. The magnetic entropy SmS_m is calculated from the integration of Cm/TC_m/T versus TT. The estimated entropy SmS_m increases and saturates to the maximum value of about 14.83 J/mol K (= Rln⁡6R\ln6), expected for an S=5/2S=5/2 system.

The inset of Figure 5 shows the plot of Cm/TC_m/T versus T2T^2. We have compared the data with the linear and exponential curves. The data do not follow the exponential behavior, ruling out the existence of a spin-gap in the ground state. A very small upturn in Cm/TC_m/T versus T2T^2 plot indicates that the system might be reaching a static magnetic long-range ordered state at extremely low temperatures. In Figure 5, The Cm/TC_m/T is nearly independent of TT at low temperatures (T≪J/kBT\ll J/k_B), indicating that data follows nearly linear behavior expected for the systems with gapless excitations. From the Cm/TC_m/T versus T2T^2 plot, the intercept value is found to be ≈250\approx250 mJ/mol-K2^2. It is somewhat larger than that of S=5/2S=5/2 linear chain system tetramethyl ammonium manganese trichloride (TMMC) with J/kB=−6.7J/k_B=-6.7 K (γ≈9.8\gamma\approx9.8 mJ/mol-K2^2) [27]. From the normalized magnetic heat capacity (Cm/JNkB2C_m/JNk_B^2) [6], the scaled γ\gamma value (i.e. γ/JNkB2\gamma/JNk_B^2) for Bi3_3FeMo2_2O12_{12} (S=5/2S=5/2 triangular chain) is different from TMMC (S=5/2S=5/2 linear chain). The results suggest that S=5/2S=5/2 triangular spin chain system Bi3_3FeMo2_2O12_{12} hosts the robust gapless excitations.

Cm/T versus T plot

Figure 5. (Color online) Cm/TC_m/T vs. TT plot. Inset shows Cm/TC_m/T vs. T2T^2 with the comparison of linear and exponential curves.

DFT calculations

To shed light on the electronic structures and to understand the magnetic behavior of Bi3_3FeMo2_2O12_{12}, spin-polarized DFT calculations in the local-spin density approximation (LSDA) and LSDA+UU (Hubbard UU) approach were carried out by means of a full-potential linearized muffin-tin orbital (FP-LMTO) method [28, 29] as implemented in the RSPt code [30]. We have considered on-site Coulomb interaction U=2U=2 eV combined with Hund’s exchange JH=0.8J_H=0.8 eV within the fully rotationally invariant LSDA+UU approach [31] to treat the electronic correlation effects of Fe-dd states. Such choices of UU are guided by a previous report on a correlated oxide containing high-spin Fe3+^{3+} ions [32]. From both LSDA and LSDA+UU calculations, the lowest energy state is identified for the pattern with the antiferromagnetic couplings between the 1st^{st} NN and 2nd^{nd} NN Fe-spins in the isolated zig-zag-chain.

J1J_1J2J_2
Fe-Fe distance3.8 Å5.3 Å
Exchange pathFe-O1-O1-Fe (Fe-O1-O1 = 128.5° & O1-O1-Fe = 71.9°)Fe-O2-O2-Fe (Fe-O2-O2 = 111.8° & O2-O2-Fe = 111.8°)
LSDA+UU-1.52 K-1.66 K

Table I. The details of exchange coupling strengths using LSDA approach with UU

The computed total and orbital-decomposed density of states (DOS) in this lowest-energy magnetic state as obtained from LSDA+UU are shown in Figure 6(a). The DOS of Fe-3d3d clearly shows that the majority spin states are fully filled up, and the minority spin states are empty (see Figure 6(b)). This is consistent with the picture of the high-spin ground state of Fe3+^{3+} ions (3d53d^5). The Mo-dd states (see Figure 6(c)) are found to be completely empty, indicating the nominal d0d^0 non-magnetic state of these ions. The O-pp states are also delocalized and spread in the entire valence band. The insulating gap of 1.7 eV is found between O-pp and the Fe-dd states, making this system a member of the charge-transfer insulator in the Zaanen, Sawatzky, and Allen (ZSA) scheme [33, 34].

Total and orbital-decomposed density of states

Figure 6. Total and orbital-decomposed density of states (DOS) for the lowest energy magnetic state. Fermi energy is set to zero.

We next estimated the various magnetic exchange couplings (J1J_1 and J2J_2 as marked in Figure 1) based on the converged lowest energy magnetic state. Here we employed the formalism of Ref. [35], where the total converged energies of the magnetic system are mapped onto a Heisenberg Hamiltonian, and magnetic force theorem [36, 37] is applied to extract the inter-atomic magnetic exchange interactions. Our calculations, as summarized in table I, reveal that both J1J_1 and J2J_2 are antiferromagnetic. The J1J_1 and J2J_2 exchange interactions are mediated through Fe-O-O-Fe super-exchange paths with varied Fe-Fe distances (3.79 Å and 5.25 Å), as mentioned in tableI. Interestingly, the magnitudes of both the exchanges come out to be almost equal (J2/J1≈1.1J_2/J_1 \approx1.1) although the corresponding Fe-Fe distances are very different. We also note that such a conclusion is very robust and independent of the adopted methods and independent of UU choice within the LSDA+UU approach.

The highly delocalized nature of O-pp orbitals promotes such super-super exchange interaction. Interestingly, the Fe-O-O angles that connect the two 2nd2^{nd} NN Fe are equal while these angles are different for 1st1^{st} NN. The peculiar character of the crystal symmetry is responsible for the reason of having J1J_1 and J2J_2 to be nearly equal, despite having a difference in the Fe-Fe bond distances. The nature of the exchange could also be qualitatively understood within the framework of the extended Kugel-Khomskii model [38, 39]. It is well established that half-filled orbitals promote antiferromagnetic super-exchange since virtual hopping between Fe orbitals is allowed only if they possess anti-parallel alignments. Importantly, the antiferromagnetic nature, together with the comparable exchange interactions (J1≈J2J_1 \approx J_2), causes strong spin-frustration in the zig-zag-shaped triangular Fe3+^{3+} chains. The estimated θCW∼−37.5\theta_{CW} \sim-37.5 K agrees very well with the experimental value of −40-40 K, providing further credence to the theoretically computed values of the exchange interactions. The ratio of the inter- to intra-chain magnetic exchange coupling J′/J1J'/J_1 is estimated at about 0.01. Thus, we can conclude that the present system is an example of a nearly isolated S=5/2S = 5/2 triangular spin chain.

Discussion

Triangular antiferromagnets offer a promising ground for realizing the unusual states of matter. In S=5/2S = 5/2 triangular systems, a few 2D magnets were investigated [FeO2_2 (A=Cu, Li, and Na)] [40, 41], however, the quantum ground state without magnetic LRO has not been identified. Achieving the disordered quantum state in S=5/2S = 5/2 systems probably requires a material with ideal one-dimensionality. Our experimental observations on S=5/2S = 5/2 triangular chain material Bi3_3FeMo2_2O12_{12} (J2/J1≈1.1J_2/J_1 \approx1.1 and J′/J≈0.01J'/J \approx0.01) revealed that it exhibits large magnetic frustration (f>200f > 200). Besides, the observation of the linear behavior of Cm(T)C_m(T) suggests the presence of gapless excitations. Strikingly, the physics of S=5/2S = 5/2 triangular chain behavior is entirely different from that of the S=1/2S = 1/2 triangular chain, where one can expect a robust spin-gap ground state [16]. As an example, S=1/2S = 1/2 zig-zag chain system Sr0.9_{0.9}Ca0.1_{0.1}CuO2_2 has shown the spin-gap ground state due to the randomness

a Heisenberg Hamiltonian, and magnetic force theorem [36, 37] is applied to extract the inter-atomic magnetic exchange interactions. Our calculations, as summarized in table I, reveal that both J1J_1 and J2J_2 are antiferromagnetic. The J1J_1 and J2J_2 exchange interactions are mediated through Fe-O-O-Fe super-exchange paths with varied Fe-Fe distances (3.79 Å and 5.25 Å), as mentioned in tableI. Interestingly, the magnitudes of both the exchanges come out to be almost equal (J2/J1≈1.1J_2/J_1 \approx1.1) although the corresponding Fe-Fe distances are very different. We also note that such a conclusion is very robust and independent of the adopted methods and independent of UU choice within the LSDA+UU approach.

The highly delocalized nature of O-pp orbitals promotes such super-super exchange interaction. Interestingly, the Fe-O-O angles that connect the two 2nd2^{nd} NN Fe are equal while these angles are different for 1st1^{st} NN. The peculiar character of the crystal symmetry is responsible for the reason of having J1J_1 and J2J_2 to be nearly equal, despite having a difference in the Fe-Fe bond distances. The nature of the exchange could also be qualitatively understood within the framework of the extended Kugel-Khomskii model [38, 39]. It is well established that half-filled orbitals promote antiferromagnetic super-exchange since virtual hopping between Fe orbitals is allowed only if they possess anti-parallel alignments. Importantly, the antiferromagnetic nature, together with the comparable exchange interactions (J1≈J2J_1 \approx J_2), causes strong spin-frustration in the zig-zag-shaped triangular Fe3+^{3+} chains. The estimated θCW∼−37.5\theta_{CW} \sim-37.5 K agrees very well with the experimental value of −40-40 K, providing further credence to the theoretically computed values of the exchange interactions. The ratio of the inter- to intra-chain magnetic exchange coupling J′/J1J'/J_1 is estimated at about 0.01. Thus, we can conclude that the present system is an example of a nearly isolated S=5/2S = 5/2 triangular spin chain.

Discussion

Triangular antiferromagnets offer a promising ground for realizing the unusual states of matter. In S=5/2S = 5/2 triangular systems, a few 2D magnets were investigated [FeO2_2 (A=Cu, Li, and Na)] [40, 41], however, the quantum ground state without magnetic LRO has not been identified. Achieving the disordered quantum state in S=5/2S = 5/2 systems probably requires a material with ideal one-dimensionality. Our experimental observations on S=5/2S = 5/2 triangular chain material Bi3_3FeMo2_2O12_{12} (J2/J1≈1.1J_2/J_1 \approx1.1 and J′/J≈0.01J'/J \approx0.01) revealed that it exhibits large magnetic frustration (f>200f > 200). Besides, the observation of the linear behavior of Cm(T)C_m(T) suggests the presence of gapless excitations. Strikingly, the physics of S=5/2S = 5/2 triangular chain behavior is entirely different from that of the S=1/2S = 1/2 triangular chain, where one can expect a robust spin-gap ground state [16]. As an example, S=1/2S = 1/2 zig-zag chain system Sr0.9_{0.9}Ca0.1_{0.1}CuO2_2 has shown the spin-gap ground state due to the randomness [42] All the results support that the titled compound might be a possible candidate for gapless spin liquid. Muon spin relaxation and inelastic neutron scattering experiments in sub-Kelvin temperature may shed microscopic insights into the ground state and spin dynamics of the titled material.

Conclusion

Our investigation reveals that the titled compound Bi3FeMo2O12\mathrm{Bi}_{3}\mathrm{Fe}\mathrm{Mo}_{2}\mathrm{O}_{12} is a nearly ideal quasi-one-dimensional S=5/2S = 5/2 triangular chain system with a small anisotropy (J2/J1≈1.1J_{2}/J_{1} \approx1.1). The presence of strong magnetic frustration and negligible inter-chain interactions preclude magnetic LRO down to 200 mK. Cm(T)C_{m}(T) data show a linear behavior reflecting the gapless excitations in the ground state. These results will stimulate both theoretical and experimental interests to examine whether our S=5/2S = 5/2 triangular chain Bi3FeMo2O12\mathrm{Bi}_{3}\mathrm{Fe}\mathrm{Mo}_{2}\mathrm{O}_{12} can host the spin liquid ground state.

Acknowledgments: B. K thanks DST INSPIRE faculty award-2014 scheme. We thank Prof. P. L. Paulose and Dr. R. Kumar for their support in magnetic measuemets. AKM thanks IIT Tirupati and DST-SERB (Grant No.: ECR/2017/001903), Govt. of India, for providing research grants. The work at SNU were supported by NRF(2019RIA2C2090648 and 2019M3E4A1080227). PK acknowledges the funding by the Science and Engineering Research Board, and Department of Science and Technology, India through Research Grants.

References

  1. [1]A. N. Vasiliev, O. Volkova, E. Zvereva, and M. Markina, Milestones of low-D quantum magnetism, Npj Quantum Mater. 3, 1 (2018).DOI
  2. [2]A. N. Vasiliev, O. S. Volkova, E. A. Zvereva, and M. M. Markina, Low-Dimensional Magnetism, CRC Press-1st ed., Hardcover Book (2019).
  3. [3]L. Balents, Spin liquids in frustrated magnets, Nature 464, 199 (2010).
  4. [4]C. Broholm, R. J. Cava, S. A. Kivelson, D. G. Nocera, M. R. Norman, and T. Senthil, Quantum spin liquids, Science 367, 6475 (2020).
  5. [5]N. D. Mermin and H. Wagner, Absence of Ferromagnetism or Antiferromagnetism in One or Two-Dimensional Isotropic Heisenberg Models, Phys. Rev. Lett. 17, 1133 (1966).
  6. [6]D. C. Johnston, R. K. Kremer, M. Troyer, X. Wang, A. Klümper, S. L. Budko, A. F. Panchula, and P. C. Canfield, Thermodynamics of spin S = 1/2 antiferromagnetic uniform and alternating-exchange Heisenberg chains, Phys. Rev. B 61, 9558 (2000).DOI
  7. [7]J. Schlapka, P. Wohlfeld, K. J. Zhou, M. Mourigal, M. W. Haverkort, V. N. Strocov, L. Hozoï, C. Monney, S. Nishimoto, S. Singh, and A. Revcolevschi, Spin–orbital separation in the quasi-one-dimensional Mott insulator Sr2CuO3, Nature 485, 82 (2012).
  8. [8]M. Mourigal, M. Enderle, A. Klöpperpieper, J. S. Caux, A. Stunault, and H. W. Rönnow, Fractional spin excitations in the quantum Heisenberg antiferromagnetic chain, Nature Physics 9, 435 (2013).DOI
  9. [9]C. K. Majumdar, and D. K. Ghosh, On next-nearest-neighbor interaction in linear chain. II, J. Math. Phys. 10, 1399 (1969).DOI
  10. [10]S. Lebernegg, O. Janson, I. Rousochatzakis, S. Nishimoto, H. Rosner, and A. A. Tsirlin, Frustrated spin chain physics near the Majumdar-Ghosh point in szenicite Cu3(MoO4)(OH)4, Phys. Rev. B 95, 035145 (2017).DOI
  11. [11]A. K. Bera, B. Lake, W. D. Stein, and S. Zander, Magnetic correlations of the quasi-one-dimensional half-integer spin-chain antiferromagnets SrM2V2O8 (M = Co, Mn), Phys. Rev. B 89, 094402 (2014).DOI
  12. [12]L. D. Sanjeewa, V. O. Garlea, M. A. McGuire, C. D. McMillen, H. B. Cao, and J. W. Kolis, Structural and Magnetic Characterization of the One-dimensional S = 5/2 Antiferromagnetic Chain System SrMn(VO4)(OH), Phys. Rev. B 93, 224407 (2016).DOI
  13. [13]L. D. Sanjeewa, A. S. Sefat, M. A. Smart, M. A. McGuire, C. D. McMillen, J. W. Kolis, Synthesis, structure and magnetic properties of Ba3M2Ge4O14(M= Mn and Fe): Quasi-one-dimensional zig-zag chain compounds, Journal of Solid State Chemistry 283, 121090 (2020).
  14. [14]P. S. Berdonosov, E. S. Kuznetsova, V. A. Dolgikh, A. V. Sobolev, I. A. Presniakov, A. V. Oleniev, B. Rahaman, T. Saha-Dasgupta, K. V. Zakharov, E. A. Zvereva, O. S. Volkova, and A. N. Vasiliev, Crystal structure, physical properties, and electronic and magnetic structure of the spin S = 5/2 zig-zag chain compound Bi2Fe(SeO3)2OCl3, Inorganic Chemistry 53, 5830 (2014).
  15. [15]A. W. Sleight, and W. Jeitschko, Bi3(FeO4)(MoO4)2 and Bi3(GaO4)(MoO4)2- new compounds with scheelite related structures, Materials Research Bulletin 9, 951 (1974).
  16. [16]K. Uematsu, T. Hikihara, and H. Kawamura, Frustration-induced quantum spin liquid behavior in the S = 1/2 random-bond Heisenberg antiferromagnet on the zig-zag chain, arXiv preprint arXiv:2009.08630 (2020).DOI
  17. [17]C. Li, Z. Gao, X. Tian, J. Zhang, D. Ju, Q. Wu, W. Lu, Y. Sun, D. Cui, and X. Tao, Bulk crystal growth and characterization of the bismuth ferrite-based material Bi3FeO4(MoO4)2, Cryst Eng Comm, 21, 2508 (2019).
  18. [18]R. Nath, A. V. Mahajan, N. Büttgen, C. Kegler, A. Loidl, and J. Bobroff, study of one-dimensional nature of S = 1/2 (Sr, Ba)2Cu(PO4)2 and BaCuP2O7 via 31P NMR, Phys. Rev. B 71, 174436 (2005).DOI
  19. [19]B. Koteswarao, R. Kumar, P. Khuntia, Sayantika Bhowal, S. K. Panda, M. R. Rahman, A. V. Mahajan, I. Dasgupta, M. Baenitz, Kee Hoon Kim, and F. C. Chou, Magnetic properties and heat capacity of the three-dimensional frustrated S = 1/2 antiferromagnet PbCuTe2O6, Phys. Rev. B 90, 035141 (2014).
  20. [20]Y. Okamoto, M. Nohara, H. Aruga-Katori, and H. Takagi, Spin-Liquid State in the S = 1/2 Hyperkagome Antiferromagnet Na4Ir3O8, Phys. Rev. Lett. 99, 137207 (2007).DOI
  21. [21]M. E. Fisher, and J. C. Bonner, Linear Magnetic Chains with Anisotropic coupling, Phys. Rev. 135, A640 (1964).DOI
  22. [22]M. E. Fisher, Magnetism in one-dimensional systems-the Heisenberg model for infinite spin, American Journal of Physics 32, 343 (1964).DOI
  23. [23]L. J. De Jongh, and A. R. Miedema, Experiments on simple magnetic model systems, Advances in Physics 50, 947 (2001).
  24. [24]To match both the Cₚ data at high temperatures from 200 K to 250 K, the Cₚ data of Bi₃GaMo₂O₁₂ is multiplied with 1.09 and then used for substracting the lattice of Bi₃FeMo₂O₁₂.
  25. [25]C. Kittel, Introduction to Solid State Physics, John Wiley and Sons press - 8ᵗʰ ed., New York, NY (2004).
  26. [26]B. Koteswararao, S. K. Panda, R. Kumar, Kyongjun Yoo, A. V. Mahajan, I. Dasgupta, B. H. Chen, Kee Hoon Kim, and F. C. Chou, Observation of S = 1/2 quasi-1D magnetic and magneto-dielectric behavior in a cubic SrCuTe₂O₆, J. Phys. Condens. Matter 27, 426001 (2015).
  27. [27]W. J. M. De Jonghe, C. H. W. Swüstje, K. Kopinga, and K. Takeda, Specific heat of nearly-one-dimensional tetramethyl ammonium manganese trichloride (TMMC) and tetramethyl ammonium cadmium trichloride (TMCC), Phys. Rev. B 12, 5858 (1975).
  28. [28]O. K. Andersen, Linear methods in band theory, Phys. Rev. B 12, 3060 (1975).
  29. [29]J. M. Wills and B. R. Cooper, Synthesis of band and model Hamiltonian theory for hybridizing cerium systems, Phys. Rev. B 36, 3809 (1987).pubmed.ncbi.nlm.nih.gov/9943318
  30. [30]J. M. Wills, O. Eriksson, M. Alouni, and D. L. Price, Electronic Structure and Physical Properties of Solids: The Uses of the LMTO Method (Springer, Berlin, 2000).
  31. [31]V. I. Anisimov, I. V. Solovyev, M. A. Korotin, M. T. Czyzyk, and G. A. Sawatzky, Density-functional theory and NiO photoemission spectra, Phys. Rev. B 48, 16929 (1993).pubmed.ncbi.nlm.nih.gov/10008291
  32. [32]J. Chakraborty and I. Dasgupta, First principles study of electronic structure, magnetism and ferroelectric properties of rhombohedral AgFeO₂, J. Magn. Magn. Mater. 487, 165296 (2019).DOI
  33. [33]J. Zaanen, G. A. Sawatzky and J. W. Allen, Band gaps and electronic structure of transition metal compounds, Phys. Rev. Lett. 55, 418 (1985).pubmed.ncbi.nlm.nih.gov/10032345
  34. [34]S. Nimkar, D. D. Sarma, H. R. Krishnamurthy, and S. Ramasesha, Mean-field results of the multiple-band extended Hubbard model for the square-planar CuO₂ lattice, Phys. Rev. B 48, 7355 (1993).
  35. [36]A. Lichtenstein, M. Katsnelson, V. Antropov, and V. Gubanov, Local spin density functional approach to the theory of exchange interactions in ferromagnetic metals and alloys, J. Magn. Magn. Mater. 67, 65 (1987).
  36. [37]M. I. Katsnelson and A. I. Lichtenstein, First-principles calculations of magnetic interactions in correlated systems, Phys. Rev. B 61, 8906 (2000).DOI
  37. [38]K. I. Kugel and D. I. Khomskii, Crystal-structure and magnetic properties of substances with orbital degeneracy, Zh. Eksp. Teor. Fiz. 64, 1429 (1973).
  38. [39]K. I. Kugel and D. I. Khomskii, The Jahn-Teller effect and magnetism: transition metal compounds, Sov. Phys. Uspekhi 25, 231 (1982).
  39. [40]F. Ye, Y. Ren, Q. Huang, J. A. Fernandez-Baca, Pengcheng Dai, J. W. Lynn, and T. Kimura, Spontaneous spin-lattice coupling in the geometrically frustrated triangular lattice antiferromagnet CuFeO₂, Phys. Rev. B 73, 220404(R) (2006).
  40. [41]M. Tabuchi, K. Ado, H. Sakaebe, C. Masquelier, H. Kageyama, and O. Nakamura, Preparation of aFeO₂ (A = Li, Na) by hydrothermal method, Solid State Ionics 79, 220 (1995).DOI
  41. [42]F. Hammerath, S. Nishimoto, H.-J. Grafe, A. U. B. Wolter, V. Kataev, P. Ribeiro, C. Hess, S.-L. Drechsler, and B. Büchner, Spin Gap in the zig zag Spin-1/2 Chain Cuprate Sr₀.₉Ca₀.₁CuO₂, Phys. Rev. Lett. 107, 017203 (2011).DOI

Paper details

vlnc:W_tbquupzj
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