WorksPhysics and Astronomy
Spin-glass behavior in Shastry-Sutherland lattice of TmCuIn
Abstract
The spin-glass behavior in the ferromagnetic phase of was investigated by dc-, ac-magnetization, and non-equilibrium dynamics characterizations. Arc-melted polycrystalline compound of crystallizes in -type tetragonal structure (space group ). The temperature variation dc-magnetization exhibits a ferromagnetic behavior with Curie temperature = 32.5 K. The zero-field cooled (ZFC) and field cooled (FC) magnetic curves show a thermomagnetic irreversible behavior below . The field dependence of irreversible temperature follows Almeida-Thouless line. The frequency and ac-driven field-dependent anomalies in ac-susceptibility results indicate the existence of spin-glass state in . The time dependence of magnetization further supports the spin-glass behavior observed in . The frequency dependence of the freezing temperature in the real part of ac-susceptibility has been analyzed on the basis power-law divergence and Vogel-Fulcher law. The obtained results revealed that belongs to a ferromagnetic canonical spin-glass class of compounds.
Ternary rare-earth based RTX (R = rare-earth, T = transition metals, and X = p block elements), magnetic materials have been intensively studied because of their exotic physical properties1–4. Among them, the RTX series of compounds are of interest due to their diversity of magnetic properties due to the crystal structure. Till date, RTX compounds are mainly reported to form into two types of crystal structure, namely tetragonal and orthorhombic2,5,6. The tetragonal structure of RTX with space group P4/mbm is well known as the Shastry–Sutherland lattice systems. The arrangement of rare earth ions as triangles in the Shastry–Sutherland lattice favours the occurrence of magnetic frustration7–9. It is reported that the tetragonal structure of GdAuCd exhibits ferromagnetic spin–glass anomalies10.
The geometrically frustrated lattice in a magnetic system can exhibit spin–glass behavior. Since last few decades, spin–glass studies remain an attractive topic in magnetism[12, 13] [11]. Tm-based intermetallic compounds have been paid significant attention in the search for low temperature magnetic properties due to their relatively low transition temperature. In this paper, the magnetic properties of TmCuIn have systematically been investigated. The obtained ferromagnetic spin–glass behavior is reported.
A polycrystalline sample of TmCuIn was prepared by the standard arc-melting technique on a water-cooled copper hearth under high purity Argon atmosphere. The high-purity (≥ 99.9 wt %) elements with the stoichiometric 2:2:1 ratio was melted. The ingot was flipped and remelted several times to improve the homogeneity and reaction among the constituents. The weight loss of the melted ingots was less than 0.5 wt.%. The melted ingot of TmCuIn was then wrapped in tantalum foil and was kept in a highly evacuated sealed quartz tube. The sample was annealed at 973 K for a week. The crystal structure of the annealed sample was confirmed from the powder x-ray diffraction (XRD) pattern using a Rigaku Diffractometer with copper K radiation. The dc–magnetization and ac–magnetic susceptibility were measured by using a Dynacool–Physical Properties Measurement System (PPMS), from Quantum–Design, San Diego.
The powder XRD data at room temperature collected on the polycrystalline sample of TmCuIn were analyzed by Rietveld structural refinement using FULL-
| a | 7.399(2) Å | |||
| b | 7.399(2) Å | |||
| c | 3.697(2) Å | |||
| V | 202.368(2) Å | |||
| Atomic coordinates for TmCuIn | ||||
| Atom | Wyckoff | |||
| Tm | 0.1747(1) | 0.6747(1) | 1/2 | |
| Cu | 0.6212(1) | 0.1212(1) | 0 | |
| In | 0 | 0 | 0 |
Table 1. The lattice parameters and unit cell volume of TmCuIn compound obtained from Rietveld refinement of XRD patterns for the tetragonal phase along with the atomic coordinate positions.
PROF software[14, 15]. Figure 1 shows the XRD pattern of TmCuIn along with the Rietveld fitted profile. The Rietveld refinement of the XRD data reveals that TmCuIn shows single phase compound of MoFeB–type tetragonal crystal structure with space group P4/mbm. The obtained crystallographic parameters from the Rietveld refinement are listed in Table I. The observed crystallography parameters are very close to the values previously reported[16] [5]. There are two Tm–Tm bond distances in the structure. The shortest one measures 0.3656 nm (shown as dotted line) while the longer one measures 0.3864 nm (shown as solid line).
Figure 1. Room temperature powder X–ray diffraction patterns along with the refinement profile of TmCuIn. Red symbols represent the experimental data and the black line represents the calculated data. The difference between experimental and calculated data is shown as a blue line. A set of vertical bars represent the Bragg peak positions of the tetragonal MoFeB type structure with the space group P4/mbm. Inset: The schematic representation of the tetragonal crystal structure of TmCuIn. The wine colour balls are for Tm, blue colour are for Cu and green colour are for In.
The zero field cooled (ZFC) and field cooled (FC) dc–magnetization as function of temperature for different external applied magnetic fields starting from 0.25 kOe to 5 kOe of TmCuIn are shown in Figure 2. FC magnetization shows a typical ferromagnetic behavior with transition temperature = 32.5 K. The obtained value is in good agreement with the earlier report[17] [6]. The ZFC and FC curves start to diverge below a certain temperature, called the irreversible temperature (). It is seen that strongly depends on the applied magnetic field. The experimentally obtained for different applied magnetic fields was estimated from the ZFC and FC magnetization. Inset of Figure 2 shows the – phase diagram by plotting of verses . This – phase diagram follows the Almeida–Thouless line for strong irreversibility characteristics[18, 19]. The results of dc–magnetization and the Almeida–Thouless line have provided indications for the presence of spin–glass behavior in the compound.
Figure 2. Main panel: Temperature dependence of dc–magnetization of TmCuIn under different magnetic field. Inset; versus plot following the Almeida–Thouless law.
Spin–glass behavior may also be characterized by the magnetic relaxation method. The relaxation process can be measured by both ZFC and FC conditions. In this present case, it is measured under ZFC condition. Here, the sample was cooled without external magnetic field from the paramagnetic region to 6 K (below the freezing temperature). After reaching the desired temperature, a waiting time of 60 sec was observed and the time evolution of the magnetization [] was recorded under a tiny applied magnetic field. Figure 3 shows the time dependence of magnetization measured at 6 K in ZFC condition under the field value of 500 Oe. As seen in Figure 3, the magnetization increases slowly with time and does not saturate even after a time of 10000 s. The time dependence of the magnetization is explained by the standard stretched exponential function, as follows[20, 21].
Figure 3. (a): The time dependence of relaxation magnetization of TmCuIn measured at 6 K after zero applied magnetic field and measured under a magnetic field of 500 Oe. The red curve is the fit of the superposition of a stretched exponential expression and a constant term, see Eq. (1).
where is the spontaneous magnetization and is related with the glassy component of magnetization. and are the characteristic parameters for relaxation time constant, and stretching exponent respectively. The LSQ fitted line of Eq. (1) to the experimental data in Figure 3 yielded the parameters emu/g, emu/g, the relaxation time sec and . For a typical spin–glass system, the value of lies between 0 and 1[22, 23]. As per earlier reports, our result supports that TmCuIn exhibits a spin–glass state.
The ac–magnetic susceptibility is an important characterization to confirm and categorize spin–glass systems. Figure 4(a) and (Figure 4b) show the temperature dependence of the real () and imaginary () parts of the ac–magnetic susceptibility for different frequencies, respectively, where the dc–magnetic field was zero and ac–driven field () = 3 Oe. As seen in Figure 4(a), shows a pronounced peak around the spin–glass transition temperature. The peak temperature is known as the spin freezing temperature (). It is observed that slowly shifts to higher temperature with increasing frequency. Similarly, a frequency dependent peak is also observed in . The magnitude of the peak of increases with increasing the frequency, which is related to the dissipation process in the system. These results further confirm the spin–glass behavior in the system.
Figure 4. (a): The temperature dependence of the real parts of the ac–magnetic susceptibility measured at different frequencies in an applied ac–magnetic field of 3 Oe. (b): The temperature dependence of the imaginary parts of the ac magnetic susceptibility measured at different frequencies in an applied ac–magnetic field of 3 Oe. (c): The temperature dependence of the real part of the ac–magnetic susceptibility measured at different ac–driven field in an applied frequency 1000 Hz. (d): The temperature dependence of the imaginary part of the ac–magnetic susceptibility measured at different ac–driven fields () in an applied frequency of 1000 Hz.
To further explore the spin–glass state in TmCuIn, temperature dependent ac–susceptibility was measured under various ac–driven field (). Figure 4(c) and (Figure 4d) show the and of the ac–magnetic susceptibility for different and at a constant frequency of 1000 Hz with zero dc–magnetic field respectively. As shown in Figure 4(c) and (Figure 4d), the position of the peak in and depend on . Both and shifts to low temperature and the magnitude increases with increasing . These observations also confirm the ferromagnetic spin–glass behavior in TmCuIn.
The ac–magnetic susceptibility results confirm that TmCuIn follows spin–glass type transition with a freezing temperature K. In a typical spin–glass system, the frequency dependence of is used to compare the relative shift in freezing temperature per decade of frequency by the following expression[23]:
Here, , and with Hz and Hz. It is found a value of for TmCuIn, which lies typically in the range of a canonical spin glass system[23]. This result revealed that a suitable characterization of TmCuIn is based on the canonical spin–glass state.
The frequency dependence of for TmCuIn follows the conventional power–law divergence of a critical slowing down and is given by the relation[12, 24] [11];
where is the relaxation time corresponding to the measured frequency () and is a characteristic relaxation time of single spin flip. is known as the spin–glass temperature as frequency tends to zero. is the critical exponent of correlation length, and . is known as the dynamic critical exponent and is an important parameter of the spin–glass system. The critical exponent lies between 4 and 12 for a typical spin–glass system. Eq. (3) can be written as:
where . The frequency dependence of obtained from is plotted as log–log of inverse frequency against and is shown in Figure 5(a). The intercept and slope of the linear fit to – are used to calculate the value of and respectively. From the best fit, s and are obtained. The value of supports, the observation of spin–glass system in TmCuIn. The obtained value of for TmCuIn is close to the typical value of – s for canonical spin–glass systems.
Figure 5. (a): vs. plot for the frequency dependence of freezing temperature, where the reduced temperature . The solid line represents the fitting of power-law divergence to experimental data. (b) The frequency dependence of freezing temperature plotted as vs. . The solid straight line represents the fitting of Vogel-Fulcher law to experimental data.
A spin–glass system is also categorised using the dynamical Vogel–Fulcher scaling law[12, 25]. The frequency dependence of from is described by:
where where stand for the Boltzmann constant, is the characteristic attempt frequency, is thermal activation energy, and is known as the Vogel–Fulcher temperature. To conveniently fit the data, Eq. (5) can be rewritten as:
According to this equation, a plot of vs. is made. The slope and intercept of the linear fit are used to estimate and respectively. The best fit yielded a value of K and K. The ratio of is less than 1 confirming the system is in canonical spin–glass behavior. However, is generally found to be relatively large in cluster spin–glass state. This result revealed that TmCuIn exhibits ferromagnetic canonical spin–glass state for the geometrical frustration in the crystal structure.
A Summary
In summary, we have successfully prepared a polycrystalline sample of the intermetallic compound TmCuIn, which crystallise in the MoBFe – type of tetragonal structure. The spin–glass behavior in the ferromagnet found below its characteristic freezing temperature K, has been confirmed by dc–magnetization and ac–susceptibility measurements. The spin–glass behavior in TmCuIn has also been suggested by the magnetic relaxation behavior. The canonical spin–glass state in TmCuIn was suggested from the experimental data of frequency dependence of in ac–susceptibility. The experiment results are discussed with the theoretical model of power divergence method and Vogel–Fulcher law. This observation revealed that the investigated compound exhibits ferromagnetic canonical spin–glass state due to the geometrically frustration in the structure likely due to the formation of distorted triangles by the magnetic rare-earth ions.
Acknowledgements
This work is supported by Global Excellence and Stature (UJ-GES) fellowship, University of Johannesburg, South Africa. DFR thanks OWSD and SIDA for the fellowship towards PhD studies. AMS thanks the URC/FRC of UJ for assistance of financial support.
DATA AVAILABILITY
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
References
- [12]Mydosh, J. A. (1993). Spin Glasses. In Taylor and Francis (p. London).
- [13]Young, A. P. (1997). Spin Glasses and Random Fields. In World Scientific (p. Singapore).
- [14]14 H. M. Rietveld, J. Appl. Crystallogr., 2 (1969) 65.
- [15]Rodriguez-Carvajal, J. (2017). Fullprof Suite. http://www.ill.eu/sites/fullprof/ill.eu/sites/fullprof
- [16]16 Y. M. Kalychak, V.I. Zaremba, V.M. Baranyak, P. Y. Zavalii, V. A. Bruskov, L. V. Sysa L.V., and O. V. Dmytrakh, Izv. Akad. Nauk SSSR, Neorg. Mater., 26 (1990) 74–76.
- [17]17 Y. Zhang, Y. Yang, X. Xu, L. Hou, Z. Ren, X. Li, and G. Wilde, J. Phys. D: Appl. Phys. 49 (2016) 145002.
- [18]18 J. R. L. de Almeida and D. J. Thouless, J. Phys. A 11 (1978) 983.
- [19]19 M. Gabay and G. Toulouse, Phys. Rev. Lett. 47 (1981) 201.
- [20]20 S. Ghara, B. G. Jeon, K. Yoo, K. H. Kim, and A. Sundaresan, Phys. Rev. B 90 024413 (2014).
- [21]21 A. Bhattacharyya, S. Giri, and S. Majumdar, Phys. Rev. B 83 134427 (2011).
- [22]22 D. Chu, G. G. Kenning, and R. Orbach, Phys. Rev. Lett. 72 3270 (1994).
- [23]Mydosh, J. A. (1993). Spin Glasses: An Experimental Introduction. In Taylor and Francis (p. London).
- [24]24 P. C. Hohenberg and B. I. Halperin, Rev. Mod. Phys. 49 (1977) 435.
- [25]25 J. Souletie and J. L. Tholence, Phys. Rev. B 32 (1985) 516.