betaFeedback

WorksPhysics and Astronomy

Magnetic field induced modification of a first-order ferromagnetic transition in Eu2In

Abstract

We present a comprehensive study of the temperature- and magnetic-field-dependent magnetization, specific heat, and local crystal structure across the first-order ferromagnetic-paramagnetic transition in Eu2_2In. Anomalies in the magnetocaloric response are observed near H≈25H \approx 25~kOe, including changes in field scaling of magnetic entropy, local entropy exponent, and universal master curve, which suggest an apparent weakening of the first-order character of the transition. However, quantitative analysis of the magnetocaloric parameters together with modified Arrott plots demonstrates that the transition remains first order up to at least 70~kOe. Specific-heat measurements reveal a field-induced splitting of the sharp zero-field anomaly into a doublet, providing a natural explanation for the change in the magnetocaloric response. Magnetic field dependent extended x-ray absorption fine structure (EXAFS) measurements show no detectable field-induced changes in the local coordination environment of Eu. We therefore attribute these observations to a magnetic field induced two-step transition process in Eu2_2In.

Introduction

A fundamental understanding of the thermodynamic nature of phase transitions is essential for elucidating the non-trivial responses of materials to applied magnetic fields, which underpin a wide range of magnetofunctional properties [1, 2]. In particular, first-order magnetic phase transitions (FOMPTs), characterized by their discontinuous nature, often exhibit strong non-linear responses to external stimuli and are therefore of considerable interest for functional applications [3, 4, 5]. Giant magnetocaloric effects (GMCE), for example, are typically observed in systems undergoing discontinuous magnetic or magnetostructural transitions [5, 6, 7, 8]. A prototypical case is Gd5(Si2Ge2)Gd_5(Si_2Ge_2), which exhibits a pronounced GMCE near room temperature associated with a first-order phase transition (FOPT) [5, 9, 10]. In many materials, such transitions originate from strong coupling between magnetic order and lattice degrees of freedom, that are inherently accompanied by thermal hysteresis. Although extensive efforts over the past decades have focused on understanding the thermodynamic processes occurring in the vicinity of FOPTs [11, 12, 13, 14, 15], hysteresis associated with the transitions remains a major challenge, as it leads to energy dissipation, reduces refrigeration efficiency, and limits long-term cycling stability in solid-state caloric applications [16, 17].

In this context, R2R_2In (R = rare earth) compounds have recently attracted particular attention [18], as certain members of this family exhibit large magnetic entropy changes (ΔSM\Delta S_\mathrm{M}) under modest magnetic fields, negligible thermal hysteresis, and minimal structural distortion at the first-order ferromagnetic (FM) to paramagnetic (PM) transition—characteristics highly desirable for practical magnetocaloric applications [19, 20, 21, 22, 23, 24, 25]. Within the R2R_2In family, compounds with R = Eu and Pr exhibit first-order transitions [19, 20], R = Nd displays a borderline behavior between first- and second-order transitions [21, 22, 23], while the remaining members undergo either second-order or no magnetic transitions [18, 25, 26]. R2R_2In compounds crystallize in a Co2_2Si-type orthorhombic structure for R = Eu and Yb [19, 27], whereas all other members adopt the Ni2_2In-type hexagonal structure [18, 28], indicating no direct connection between the crystal structure and the order of the magnetic transition in these compounds. However, the R = Eu and Yb members host divalent rare-earth ions [27, 29, 30], whereas in all other R2R_2In compounds the rare-earth elements occur in the trivalent state, suggesting an intricate correlation between the valence state and the crystal structure of these compounds.

In this family, Eu2Eu_2In exhibits the highest magnetic entropy change of ≈−28\approx-28 J/kg-K for ΔH=20\Delta H = 20 kOe, with negligible thermal hysteresis (0.1 K) and a minimal change of ≈0.1%\approx0.1\% in the unit cell volume across its first-order FM–PM transition at TC=55T_\mathrm{C} = 55 K [19]. Theoretical calculations demonstrate that the strong hybridization between Eu-5d and In-5p states near the Fermi level gives rise to long-range exchange interactions between Eu atoms, thereby triggering the ferromagnetic transition, which in turn underlies the large magnetic entropy change at low magnetic fields in this compound [19]. Ab initio calculations by Tapia et al. attribute the discontinuous phase transition and the associated large latent heat in Eu2Eu_2In to a purely electronic mechanism arising from the divalent nature of the Eu ions in the compound [31]. Subsequently, Alho et al. employed a model Hamiltonian—independent of the details of the electronic structure—that incorporates magnetic exchange and magnetoelastic interaction terms, and successfully reproduces the magnetic-field dependence of TcT_c as well as the temperature dependence of the magnetization and magnetocaloric properties [32]. Furthermore, by analyzing the evolution of mathematical critical points in the magnetic free energy as a function of magnetization with respect to temperature and applied field, the authors identified a critical magnetic field Hc≈25H_c \approx25 kOe, above which the transition in Eu2InEu_2In thermodynamically changes from first order to second order [32]. This theoretically predicted change in the order of the magnetic transition is of particular relevance for magnetocaloric applications, since a reduction in the discontinuity and sharpness of the transition at higher fields can markedly influence the magnetocaloric performance. Nevertheless, no experimental investigations have yet been conducted to verify this proposed field-induced crossover in Eu2InEu_2In.

In this study, we report a detailed investigation of the temperature- and field-dependent magnetization, specific heat, and local crystal structure across the first-order FM–PM transition in Eu2InEu_2In. Distinct anomalies are identified in the magnetocaloric response, including the maximum magnetic entropy change (ΔSMmax⁡\Delta S_{\mathrm{M}}^{\max}), the local power exponent (nn), and the universal scaling behavior, around H∼25H \sim25 kOe—the crossover field predicted in Ref. [32]. However, a quantitative analysis of the extracted parameters, together with the modified Arrott plot analysis, demonstrates that the transition retains its first-order nature up to the highest applied field of 70 kOe. The specific heat measurements reveal that the sharp zero-field anomaly evolves into a well-defined doublet with increasing magnetic field, while extended x-ray absorption fine structure (EXAFS) measurements show no detectable field-induced modifications in the local coordination environment within the experimental resolution. This behavior is discussed in terms of the possible emergence of a field-driven two-step process associated with the transition.

Experimental

Polycrystalline Eu2InEu_2In was synthesized from stoichiometric amounts of high-purity indium (99.9995%, Alfa Aesar) and europium metal (99.995%, Materials Preparation Center, Ames National Laboratory). The elements (total mass ∼3\sim3 g) were loaded into a tantalum (Ta) crucible and sealed by arc welding under an ultrapure argon atmosphere. Prior to sample loading, the Ta crucible was degassed under high vacuum (∼10−6\sim10^{-6} Torr) at 1800 °C for 2 h. The Ta crucible was then sealed in a quartz tube under high-purity He gas to prevent oxidation of the crucible at elevated temperatures. The sample was heated to 900 °C for 12 h in a resistive furnace, flipped, and reheated three times to improve homogeneity, and then annealed at 650 °C for 24 h. After furnace cooling, the sample was extracted from the crucible and handled in an Ar-filled glovebox due to its high reactivity with air. Room-temperature powder x-ray diffraction (XRD) data were collected using a Rigaku diffractometer with Mo KαK_\alpha radiation [33]. For the XRD measurements, the sample was ground using a mortar and pestle and sieved through a 36 μm mesh inside an Ar-filled glovebox. The repowered sample was mounted in a shallow circular recess on a flat glass holder and sealed between two Kapton tapes inside the glovebox. The Rietveld refinement of the XRD data was performed using the FullProf Suite, with a pseudo-Voigt peak shape and linear interpolation between background points.

Temperature- and field-dependent magnetization measurements were performed using a superconducting quantum interference device (SQUID) magnetometer (MPMS XL-7, Quantum Design, USA). The sample was loaded into an airtight holder inside the glovebox to prevent air exposure. Magnetization isotherms at successive temperatures were recorded by increasing the magnetic field from 0 to 70 kOe in a stable (no-overshoot) mode, followed by decreasing the field from 70 kOe to 500 Oe linearly and from 500 Oe to 0 Oe in an oscillatory mode before proceeding to the next temperature, taking into account the low coercivity of the sample (∼50\sim50 Oe). Specific-heat measurements were carried out using a Physical Property Measurement System (PPMS, Quantum Design, Inc.) employing both the standard 2τ2\tau relaxation technique and the long heat-pulse method [34, 35].

Temperature- and field-dependent x-ray absorption spectroscopy (XAS) measurements at the Eu L3L_3 edge were performed at the PIPOXS beamline (ID2A) of the Cornell High Energy Synchrotron Source (CHESS) using the NJIT 10 T superconducting Oxford Spectromag4000-10 magnet system. Eu2InEu_2In powder was ground and homogenized with boron nitride (BN) in a fixed ratio and pressed into a 13 mm-diameter pellet inside an inert-atmosphere glovebox. The pellet was sealed between two Kapton tapes to prevent air exposure. The sample was cooled in He vapor. At each temperature and magnetic field, four spectra were collected and averaged. Energy calibration was performed using a simultaneously measured Eu metal foil reference to correct for possible energy shifts between scans [36, 37]. X-ray absorption near-edge structure (XANES) and EXAFS data reduction and analysis were performed using Athena and Artemis software packages (Demeter suite) [38]. Background subtraction and normalization were carried out in Athena, and EXAFS fitting was performed in Artemis using theoretical scattering paths calculated with FEFF6.

Results

X-ray diffraction

The room-temperature powder XRD pattern of Eu2InEu_2In is shown in Figure 1(a). Rietveld refinement confirms that Eu2InEu_2In crystallizes in the Co2SiCo_2Si-type orthorhombic structure (space group PnmaPnma), in agreement with previous reports [19, 29]. However, careful inspection reveals a small amount (∼2%\sim2\%) of a secondary Eu8In3Eu_{8}In_{3} phase, indicated by an asterisk in Figure 1(a) (not included in the refinement). This trace amount of the secondary phase in Eu2InEu_{2}In is not uncommon and has also been observed previously [27]. In the present study we focus primarily on field-induced changes in the nature of the transition in Eu2InEu_{2}In around Tc=55T_{c}=55 K; therefore, this minor secondary phase is not expected to influence the conclusions of our findings. The increasing background at low 2θ2\theta values (≲20∘\lesssim20^{\circ}) arises from the Kapton tape used to protect the sample from air exposure. The refined lattice parameters, atomic Wyckoff positions, and refinement quality

Figure 1. (a) Rietveld refinement of the room-temperature powder XRD pattern of Eu2InEu_{2}In collected using Mo KαK_{\alpha} radiation. Red symbols represent the observed intensities, and the black solid line denotes the calculated profile. Vertical tick marks indicate the Bragg positions corresponding to the orthorhombic PnmaPnma (Co2_{2}Si-type) phase. The bottom trace shows the difference between the observed and calculated patterns. An asterisk indicates the presence of a small amount (∼2%\sim2\%) of a secondary Eu8In3Eu_{8}In_{3} phase. (b) Three-dimensional representation of the crystal structure of Eu2InEu_{2}In, highlighting the two inequivalent Eu sites. (c) Temperature-dependent magnetic susceptibility (χ\chi–TT) measured in zero field cooled (ZFC), field cooled warming (FCW), and field cooled cooling (FCC) protocols at 1 kOe in the temperature stable mode. Inset (c1) shows high resolution FCC and FCW data recorded in the temperature sweep mode in the vicinity of TcT_{c}. Inset (c2) shows the Curie–Weiss fit to the FCW data between 60 to 300 K. (d) Field-dependent magnetization measured at 2 K. The inset shows an enlarged view of the low-field region.

indicators are summarized in Table 1. Figure 1(b) shows a three-dimensional representation of the Eu2_2In crystal structure, highlighting the two crystallographically inequivalent Eu sites.

Lattice parametersAtomic positionsitexyzFitting parameters
aa (Å)7.4507(7)Eu14c4c0.0322(10)0.250.7074(6)χ2=1.96\chi^{2}=1.96
bb (Å)5.5795(5)Eu24c4c0.1802(7)0.250.0715(6)Rp=5.11R_{p}=5.11
cc (Å)10.3128(10)Rwp=7.39R_{wp}=7.39
cell volume (Å3^{3})428.72(7)In4c4c0.2380(12)0.250.3926(7)

Table 1. Room-temperature structural parameters and refinement quality indicators of Eu2InEu_{2}In obtained from Rietveld refinement of powder XRD data collected using Mo KαK_{\alpha} radiation. Values in parentheses represent the standard deviations of the last significant digits.

Magnetization

The temperature-dependent magnetic susceptibility, χ=M/H\chi= M/H, of Eu2_2In measured at an applied field of H=1H = 1 kOe under zero field cooled (ZFC), field cooled warming (FCW), and field cooled cooling (FCC) protocols is shown in Figure 1(c). A sharp magnetic transition with no apparent thermal hysteresis is observed at Tc≈55T_{\mathrm{c}} \approx55 K, consistent with the previously reported first-order ferromagnetic to paramagnetic transition [19]. However, the “temperature settle” mode of the magnetization measurement is not suitable for resolving the small thermal hysteresis associated with a first-order transition. This limitation arises because even a slight overshoot of the target temperature can drive the system across the phase boundary, transforming it from one phase to the other; upon returning to the set temperature, the system may remain trapped in the transformed state. Therefore, to reliably capture the hysteretic behavior near TcT_{\mathrm{c}}, magnetization measurements were performed upon both heating and cooling using the “temperature sweep” mode with a sweep rate of 0.1 K/min, as shown in the inset (c1) of Figure 1(c). A clear thermal hysteresis of ∼0.1\sim0.1 K is observed at TcT_{\mathrm{c}}, confirming the first-order nature of the transition. Furthermore, the ZFC susceptibility was fitted to the Curie–Weiss law, χ=C/(T−Tθ)\chi= C/(T - T_{\theta}), where CC is the Curie constant and TθT_{\theta} is the Curie–Weiss temperature. The linear fit of 1/χ1/\chi versus TT over the temperature range 60–300 K, shown in the inset (c2) of Figure 1(c), yields Tθ=40.1T_{\theta} = 40.1 K and an effective magnetic moment of μeff=8.36 μB\mu_{\mathrm{eff}} = 8.36\ \mu_{\mathrm{B}}/Eu. The fact that Tθ<TcT_{\theta} < T_{\mathrm{c}} is attributed to the very sharp first-order character of the transition. The extracted μeff\mu_{\mathrm{eff}} is slightly larger than the free-ion value expected for Eu2+^{2+} (7.94 μB\mu_{\mathrm{B}}), which may indicate a small additional contribution from exchange polarization of itinerant conduction band states (with Eu 5dd character) by the localized Eu 4ff moments, rather than a purely localized 4ff response. Figure 1(d) shows the field dependence of the magnetization measured at 2 K. The magnetization does not fully saturate even up to H=70H = 70 kOe. At 70 kOe, the moment reaches ∼12.7 μB\sim12.7\ \mu_{\mathrm{B}}/f.u., which is lower than the value expected for complete ferromagnetic alignment of two Eu2+^{2+} ions (gJJ=7 μBg_J J = 7\ \mu_{\mathrm{B}}/Eu). The hysteresis loop exhibits a small coercive field of ∼50\sim50 Oe, as highlighted in the inset showing an enlarged view of the low-field region. Overall, Eu2_2In combines minimal thermal and magnetic hysteresis with a pronounced change in magnetization at TcT_{\mathrm{c}}, characteristics that are favorable for magnetocaloric applications [5, 19].

To investigate the magnetic-field-induced changes in the nature of this transition in Eu2_2In, the virgin magnetization isotherms have been recorded across TcT_{\mathrm{c}} with a temperature step of ΔT=0.5\Delta T = 0.5 K [see Fig. 10(a) of the Appendix]. The magnetic entropy change (ΔSM\Delta S_{\mathrm{M}}) is then calculated from these isotherms by employing the Maxwell thermodynamic relation [39, 40].

ΔSM(T,H)=∫0H(∂M(T,H)∂T)HdH(1)\Delta S_{\mathrm{M}}(T,H) = \int_{0}^{H} \left(\frac{\partial M(T,H)}{\partial T}\right)_{H} \mathrm{d}H \tag*{(1)}

The ΔSM(T)\Delta S_{\mathrm{M}}(T) curves for Eu2_2In are shown in Figure 2(a) for ΔH=2.5\Delta H = 2.5–70 kOe, revealing a GMCE with a maximum magnetic entropy change (ΔSMmax⁡\Delta S_{\mathrm{M}}^{\max}) of ≈−20\approx-20 J/kg-K for ΔH=20\Delta H = 20 kOe. The value of ΔSMmax⁡\Delta S_{\mathrm{M}}^{\max} increases rapidly with increasing ΔH\Delta H up to ∼25\sim25 kOe and then rises at a much slower rate at higher fields, as shown in the inset of Figure 2(a) on a log-log scale. For a second-order FM–PM transition, ΔSMmax⁡\Delta S_{\mathrm{M}}^{\max} is expected to follow a power-law dependence on the applied magnetic field, ΔSMmax⁡∝Hn\Delta S_{\mathrm{M}}^{\max} \propto H^n [41]. In Eu2_2In, the ΔSMmax⁡(H)\Delta S_{\mathrm{M}}^{\max}(H) curve deviates from this power-law behavior for H<25H < 25 kOe, as evidenced by the nonlinearity in the log-log plot presented in the inset of Figure 2(a) [see Figs. S1(a, b) of [42] for more clarity]. In contrast, an almost linear behavior is observed for H>25H > 25 kOe, suggesting a qualitative change in the magnetic response above this field. A power-law fit to the 25–70 kOe data (solid black line in the inset) yields n=0.24n = 0.24, which is lower than the theoretical value for any established universality class of FM materials [41]. This suggests that although a notable change in the magnetic behavior of the sample is observed at H∼25H \sim25 kOe, as predicted in Ref. [32], the overall nature of the phase transition does not transform into a purely second-order type even for H>25H > 25 kOe.

Figure 2. (a) Temperature-dependent magnetic entropy change (ΔSM\Delta S_{\mathrm{M}}) of Eu2_2In at different magnetic fields. The inset shows the field dependence of the maximum magnetic entropy change (ΔSMmax⁡\Delta S_{\mathrm{M}}^{\max}) on a log-log scale, where the solid black line represents the linear fit for H>25H > 25 kOe. (b) Temperature-dependent local field exponent (nn) of ΔSM\Delta S_{\mathrm{M}} at different fields. (c) Normalized ΔSM\Delta S_{\mathrm{M}} versus scaled temperature curves at different fields, with the dashed line representing the curve for H=25H = 25 kOe. The inset shows the field dependence of the vertical dispersion at θ=−5\theta= -5, where the solid black lines represent straight fits to the data below and above H=25H = 25 kOe.

Further, the ΔSM(T)\Delta S_{\mathrm{M}}(T) curves remain nearly symmetric up to H=25H = 25 kOe and then gradually become asymmetric towards the PM region at higher fields [see Figure 2(a)]. To better understand this, we present local field exponent of entropy, nn, defined as [43]

n(H,T)=d(ln⁡∣ΔSM∣)d(ln⁡H)(2)n(H,T) = \frac{\mathrm{d}(\ln|\Delta S_{\mathrm{M}}|)}{\mathrm{d}(\ln H)} \tag*{(2)}

in Figure 2(b) at various magnetic fields. Note that for a second-order FM–PM transition, n=1n = 1 in the FM state, attains a minimum at TcT_{\mathrm{c}}, given by n(Tc)=1+(1−β−1)δ−1n(T_{\mathrm{c}}) = 1 + (1-\beta^{-1})\delta^{-1}, where β\beta and δ\delta are the critical exponents, and approaches n=2n = 2 in the PM region as expected from Curie-Weiss law [41]. In contrast, n>2n > 2 signifies a first-order transition [43]. Interestingly, in Eu2_2In, nn initially increases monotonically in the vicinity of TcT_{\mathrm{c}}, attains its maximum value of n∼6n \sim6 at H=25H = 25 kOe and then decreases with further increase in the magnetic field [see Figure 2(b)]. This suggests a possible transformation from first-order to second-order character of this FM–PM phase transition in Eu2_2In at H=25H = 25 kOe. Nevertheless, n>2n > 2 persists up to 70 kOe, which indicates that the overall nature of the transition remains first-order up to 70 kOe, but its first-order character likely decreases above 25 kOe.

To further clarify this, we construct a universal master curve (UMC) of ΔSM(T)\Delta S_{\mathrm{M}}(T) data measured at different magnetic fields by plotting the normalized entropy (ΔSM/ΔSMmax⁡)(\Delta S_{\mathrm{M}}/\Delta S_{\mathrm{M}}^{\max}) versus scaled temperature (θ)(\theta), defined as

θ={−(T−Tpk)/(Tr1−Tpk);T≤Tc(T−Tpk)/(Tr2−Tpk);T>Tc(3)\theta= \begin{cases} -(T-T_{\mathrm{pk}})/(T_{\mathrm{r1}}-T_{\mathrm{pk}}); & T \leq T_{\mathrm{c}} \\ (T-T_{\mathrm{pk}})/(T_{\mathrm{r2}}-T_{\mathrm{pk}}); & T > T_{\mathrm{c}} \end{cases} \tag*{(3)}

where Tr1T_{\mathrm{r1}} and Tr2T_{\mathrm{r2}} are the two reference temperatures below and above TcT_{\mathrm{c}}, respectively, for a certain value of h=ΔSM/ΔSMmax⁡h = \Delta S_{\mathrm{M}}/\Delta S_{\mathrm{M}}^{\max}, and TpkT_{\mathrm{pk}} represents the temperature corresponding to ΔSMmax⁡\Delta S_{\mathrm{M}}^{\max} [41]. The normalized entropy versus scaled temperature for Eu2_2In is presented in Figure 2(c) at different applied magnetic fields for h=1/2h = 1/2. For a second-order FM–PM transition, curves at different fields collapse into a single universal master curve in the entire temperature range, whereas a vertical dispersion is observed for first-order transition below TcT_{\mathrm{c}} [44]. In the present case, a clear dispersion in the scaled curves is observed as a function of HH below TcT_{\mathrm{c}} [see Figure 2(c)]. Interestingly, the curves exhibit considerably larger dispersion up to H≈25H \approx25 kOe [indicated by the dashed line in Figure 2(c)], which is subsequently reduced at higher magnetic fields. To clearly demonstrate this, we plot the field dependence of the normalized entropy at θ=−5\theta= -5 in the inset of Figure 2(c). A clear decrease in the dispersion rate is observed around 25 kOe, where the two solid black lines are straight fits to the data below and above H=25H = 25 kOe. However, although at a lower rate, the scaled entropy curves still show dispersion up to 70 kOe, which further indicates only a reduction in the first-order-like character of this magnetic transition above 25 kOe, but not a complete transformation to second-order nature, analogous to the behavior of ΔSMmax⁡\Delta S_{\mathrm{M}}^{\max} and nn, discussed above.

Specific heat

The zero-field specific heat (CPC_P) of Eu2_2In, measured as a function of temperature using the 2π2\pi relaxation method, is shown in Figure 3(a). A sharp, δ\delta-like peak is clearly observed at Tc≈55T_{\mathrm{c}} \approx55 K, indicating the first-order nature of the magnetic transition. At low temperatures, the specific heat can be expressed as

Figure 3. (a) Temperature-dependent zero-field specific heat of Eu2In measured using the standard 2τ2\tau relaxation method. The solid and dashed black curves represent the best fit of the data using Eq. (5) in the temperature range 70–100 K and the extrapolation of the fitted curve down to 2 K, respectively. The dotted horizontal line represents the classical Dulong–Petit limit of CPC_P for Eu2In. The inset shows the enlarged view of the low temperature region, where the solid black line represents the best fit using Eq. (4). (b) Magnetic contribution to the specific heat, Cmag/TC_{\mathrm{mag}}/T (left axis), and the magnetic entropy, SmagS_{\mathrm{mag}} (right axis), of Eu2In. The dashed horizontal line represents the theoretical magnetic entropy Smag=Rln⁡(2J+1)S_{\mathrm{mag}} = R\ln(2J + 1) expected for Eu2+^{2+}. The double-sided black arrow indicates the magnetic entropy associated solely with the magnetic transition.

CP=γT+βT3+AT3/2,(4)C_P = \gamma T + \beta T^3 + A T^{3/2}, \tag*{(4)}

where the first, second, and third terms correspond to electronic, lattice (phonon), and ferromagnetic magnon contributions, respectively [45]. A fit to the CPC_P data using the above equation in the range ∼\sim2.6–5 K, shown by the solid black line in the inset of Figure 3(a), yields γ=9.0(7)\gamma= 9.0(7) mJ mol−1^{-1} K−2^{-2}, β=11.7(5)\beta= 11.7(5) mJ mol−1^{-1} K−4^{-4}, and A=24.1(4)A = 24.1(4) mJ mol−1^{-1} K−5/2^{-5/2}. Using this value of γ\gamma and the Debye temperature for the lattice heat capacity, we fit the specific heat data in the high-temperature paramagnetic regime (70–200 K) by following relation [46, 47]

CP=γT+9nR(TθD)3∫0θD/Tx4ex(ex−1)2 dx,(5)C_P = \gamma T + 9nR\left(\frac{T}{\theta_{\mathrm{D}}}\right)^3 \int_0^{\theta_{\mathrm{D}}/T} \frac{x^4 e^x}{(e^x-1)^2}\,\mathrm{d}x, \tag*{(5)}

where nn is the number of atoms per formula unit (three in the present case), RR is the universal gas constant, and θD\theta_{\mathrm{D}} is the Debye temperature. The best fit of the CPC_P data, shown by the solid black curve in Figure 3(a), gives θD=113±2\theta_{\mathrm{D}} = 113 \pm2 K, which is in good agreement with the value reported for Pr2_2In [48]. The dashed curve in Figure 3(a) represents the extrapolation of the fitted curve down to 2 K, while the dotted horizontal line denotes the classical Dulong–Petit limit, 3nR=74.83nR = 74.8 J mol−1^{-1} K−1^{-1}.

The magnetic contribution to the specific heat, CmagC_{\mathrm{mag}}, is obtained by subtracting the lattice contribution estimated from the nonmagnetic isostructural reference compound Yb2_2In [18], by approximating the lattice heat capacity of Eu2_2In to be equal to that of Yb2_2In (see Figure 11 of the Appendix). The resulting CmagC_{\mathrm{mag}} is then used to calculate the magnetic entropy, using

Figure 11. Temperature-dependent specific heat of Eu2In and its isostructural nonmagnetic reference compound Yb2In measured in zero magnetic field using the 2τ2\tau relaxation method. The Yb2In data are shown both as measured and after mass correction to account for the difference in atomic masses. The horizontal dashed line represents the classical Dulong–Petit limit.

Smag(T)=∫0TCmag(T)TdT.(6)S_{\mathrm{mag}}(T) = \int_{0}^{T} \frac{C_{\mathrm{mag}}(T)}{T}dT. \tag*{(6)}

The Cmag/TC_{\mathrm{mag}}/T and SmagS_{\mathrm{mag}} are shown on the left and right axes of Figure 3(b), respectively. The dashed horizontal line represents the theoretical magnetic entropy value Smag=Rln⁡(2J+1)S_{\mathrm{mag}} = R\ln(2J + 1) expected for Eu2+^{2+} with J=7/2J = 7/2. Interestingly, despite the sharp first-order δ\delta-like anomaly in CPC_P, only about 25% of the total Eu2+^{2+} magnetic entropy is released at the transition itself [as indicated by the double-sided arrow in Figure 3(b)], with the remaining entropy being recovered gradually at lower temperatures.

This implies that the magnetic entropy change ΔSM\Delta S_M in Eu2_2In could, in principle, be enhanced by approximately a factor of four if the full magnetic entropy were utilized, for example by tuning the free-energy landscape through chemical substitution while preserving the large magnetic moment of the ferromagnetic state [49].

Note that the standard 2τ2\tau relaxation method for specific heat measurements assumes that (i) the sample’s specific heat remains constant within the small temperature interval raised by a single heat pulse, and (ii) the heating and cooling cycles yield identical results for a given pulse [34, 35]. However, both the assumptions break down in the case of a sharp δ\delta-like first-order transition, leading to inaccuracies in both the position and magnitude of the anomaly [34, 35]. Therefore, to accurately capture the first-order transition in Eu2_2In, we apply a long heat pulse just below the transition temperature and analyze heating and cooling cycles separately [34, 50]. The temperature-dependent specific heat measurements performed using the long heat pulse method under various external magnetic fields are shown in Figure 4(a–Figure 4o).

Figure 4. (a–o) The temperature dependent specific heat of Eu₂In in the vicinity of magnetic transition at different magnetic fields measured in both heating and cooling modes, using the long heat pulse method. THT_{\mathrm{H}} and TLT_{\mathrm{L}} in panel (f) represent the temperatures corresponding to the HT and LT peaks, respectively, at 25 kOe.

In the absence of an external magnetic field, a sharp peak is observed in the CP(T)C_P(T) curves with a thermal hysteresis of ∼0.1\sim0.1 K [Figure 4(a)]. Interestingly, with increasing magnetic field, this peak evolves into a well-defined doublet, indicating the emergence of an additional transition. A similar field-induced splitting of the sharp zero-field CPC_P peak in Eu2_2In has also been observed in Ref. [19]. Once the two peaks are sufficiently separated (H≳20H \gtrsim20 kOe), the intensity of the low-temperature (LT) peak remains nearly invariant, whereas that of the high-temperature (HT) peak decreases monotonically with further increase in the magnetic field up to 70 kOe [see Figure 12 of the Appendix]. However, the magnetic entropy associated with the transitions remains nearly invariant with the external field (see the inset of Figure 13 of the Appendix and the discussion therein). To capture the wide temperature range at higher magnetic fields, where the transition becomes significantly broadened (H≳30H \gtrsim30 kOe), more than one heat pulse was employed. A reduction in the thermal hysteresis of the CP(T)C_P(T) curves is also observed with increasing magnetic field. It should be noted, however, that the true value of the thermal hysteresis cannot be reliably determined at fields where multiple heat pulses are required to cover the transition region.

Figure 12. Magnetic field dependence of the peak values of the low- and high-temperature anomalies in the specific heat of Eu2In.

Figure 13. Temperature dependent specific heat CP(T)C_P(T) of Eu2_2In in the vicinity of the magnetic transition measured using the long heat-pulse method at different applied magnetic fields. The black solid circles represent the CP(T)C_P(T) data of the nonmagnetic reference compound Yb2_2In. Inset shows the magnetic entropy SmagS_{\mathrm{mag}} associated with the transition as a function of magnetic field. The horizontal dashed line denotes the theoretical entropy Rln⁡(2J+1)R\ln(2J+1) expected for Eu2+^{2+} (J=7/2J=7/2).

Here it is important to emphasize that magnetization measurements reveal only a single transition, manifested as a sharp and symmetric peak in the dM/dTdM/dT curves [Figure 10(d–Figure 10f) of the Appendix]. To elucidate the origin of the two anomalies observed in the CP(T)C_P(T) curves, we compare their peak positions with the transition temperature, TcT_c extracted from magnetization measurements [Figure 10(f) of the Appendix], as shown in Figure 5. The specific-heat peak positions were determined by fitting the data using a sum of two Voigt functions [Figs. S2(a– c) of Ref. [42]. Notably, the position of the HT peak (THT_{\mathrm{H}}) closely follows TcT_{\mathrm{c}} obtained from magnetization, indicating its magnetic origin, which is further supported by the systematic reduction of its magnitude with increasing field (Figure 12 of the Appendix).

Figure 5. Magnetic-field dependence of the transition temperatures determined from magnetization and specific-heat measurements. The solid black line represents a linear fit to the Tc(H)T_{\mathrm{c}}(H) curve, extracted from the magnetization data. The inset shows the field-dependent separation between the two peaks observed in the specific heat curves.

Figure 10. (a) Virgin magnetization isotherms of Eu2_2In measured from 0 to 70 kOe at different temperatures across TcT_\mathrm{c}. (b) Normalized field derivative of the magnetization in the temperature range 57–81 K. (c) Temperature dependence of the critical field HcH_\mathrm{c}; the black solid line represents a linear fit to the data. (d) Temperature-dependent magnetization at different applied fields, extracted from the virgin magnetization isotherms. (e) Normalized temperature derivative of the magnetization at different fields. (f) Field dependence of the transition temperature; the red solid line represents a linear fit to the data.

In contrast, no corresponding feature is observed in the magnetization data near the LT peak of the CP(T)C_{\mathrm{P}}(T) curves, indicating that this anomaly is not directly associated with a magnetic phase transition. A natural interpretation is that the LT anomaly originates from a lattice-related instability that becomes progressively decoupled from the magnetic transition under an applied magnetic field, consistent with the almost field independence of its magnitude [Figure 12]. However, the systematic shift of both the LT and HT peaks to higher temperatures with increasing magnetic field argues against a purely structural transition that is entirely independent of magnetism. Instead, this behavior points to the presence of finite magnetoelastic coupling, whereby the applied magnetic field indirectly influences the lattice degrees of freedom. In this context, a field-induced structural response, such as a magnetostriction-driven effect, remains a plausible scenario.

A key observation is that the separation between the two peaks in the specific-heat curves increases rapidly with magnetic field up to ∼25\sim25 kOe and then remains nearly constant at higher fields, as shown in the inset of Figure 5, while both transitions broaden with increasing field. This decoupling of the two transitions near H∼25H \sim25 kOe may be responsible for the apparent suppression of the first-order-like character of the transition above this critical field, as inferred from the magnetization measurements discussed above. A comparison of the CPC_{\mathrm{P}} data obtained using the standard 2τ2\tau relaxation method and the long heat-pulse method at 35 kOe shows excellent reproducibility (Fig. S3 of Ref. [42]), confirming that the sharp zero-field anomaly indeed splits into two distinct features under applied magnetic field.

X-ray absorption spectroscopy

Magnetic field dependent x-ray absorption spectroscopy (XAS) measurements were performed to probe possible field-induced changes in the local crystal structural and/or electronic environment of Eu2_2In. Figures 6(a) and 6(b) show the normalized XANES spectra measured at the Eu L3_3 edge under different magnetic fields at 50 K (<Tc< T_{\mathrm{c}}) and 70 K (>Tc> T_{\mathrm{c}}), respectively. An intense feature around 6976 eV is attributed to the 2p3/2→5d2p_{3/2} \to5d transition of Eu2+^{2+} [51, 52, 53], indicating a predominantly divalent state of Eu in the sample, consistent with the magnetization measurements discussed above and with previous reports [19, 30]. More importantly, the valence state of Eu remains invariant with applied magnetic field at both temperatures. Indeed, the entire XANES line shape remains unchanged with HH, including the position of the rising-edge region, as illustrated in the insets of Figure 6(a, Figure 6b) at μ=0.7\mu= 0.7, where μ\mu denotes the normalized absorption coefficient. These results demonstrate that the electronic structure of Eu2_2In is robust against the application of an external magnetic field.

Figure 6. (a, b) Normalized XANES spectra at the Eu L3_3 edge of Eu2_2In measured under different magnetic fields at 50 and 70 K, respectively. The insets show the energy position of the rising edge at μ=0.7\mu= 0.7 as a function of magnetic field. (c, d) Magnitude of the Fourier transform of the k3χ(k)k^3\chi(k) EXAFS spectra measured at different magnetic fields at 50 and 70 K, respectively.

To further investigate the local coordination environment around Eu atoms, we performed EXAFS measurements at the Eu L3_3 edge [Figs. S4(a, b) of Ref. [42]]. Figures 6(c, d) show the magnitude of the Fourier transform of the k3χ(k)k^3\chi(k) spectra [Figs. S4(c, d) of Ref. [42]] measured at different magnetic fields for T=50T = 50 and 70 K, respectively. The χ(R)\chi(R) spectra were obtained using a Hanning window over the kk range 2.5–11.5 Å−1^{-1} with a background cutoff of Rbkg=1.8R_{\mathrm{bkg}} = 1.8 Å. The dominant feature at R∼3.5R \sim3.5 Å (phase uncorrected) arises primarily from nearest-neighbor Eu–In single-scattering paths. Upon closer inspection, this peak exhibits a noticeable asymmetry toward higher radial distance, indicating additional contributions from Eu–Eu correlations and/or longer Eu–In scattering paths. In the orthorhombic PnmaPnma structure, Eu atoms occupy two crystallographically distinct sites (Eu1 and Eu2). Each Eu atom is coordinated by five In atoms (within 4.5 Å), resulting in four distinct Eu–In bond distances. Because the two shortest Eu–In distances (one singly coordinated and one twofold coordinated) are close and cannot be reliably resolved within the present EXAFS resolution, we modeled them using an average distance ⟨Eu−In1⟩\langle\mathrm{Eu}-\mathrm{In}_1\rangle, together with independent ⟨Eu−In2⟩\langle\mathrm{Eu}-\mathrm{In}_2\rangle and ⟨Eu−In3⟩\langle\mathrm{Eu}-\mathrm{In}_3\rangle distances, where Inn\mathrm{In}_n denotes the nthn^{\mathrm{th}} nearest-neighbor In atom. To avoid overparameterization, a common Debye–Waller (DW) factor was used for all Eu–In paths.

Further, both Eu1 and Eu2 sites are coordinated by eight Eu atoms, comprising five distinct Eu–Eu distances. Since the spread among these Eu–Eu distances is below the resolution of the present EXAFS data (≲0.17\lesssim0.17 Å), we modeled them using a single average Eu–Eu bond length ⟨Eu−Eu⟩\langle\mathrm{Eu}-\mathrm{Eu}\rangle, with equal weighting of contributions from both Eu sites. The χ(R)\chi(R) spectra were first fitted at T=10T = 10 K in zero field (not shown), and the amplitude reduction factor, S02S_0^2 and energy shift, ΔE0\Delta E_0 were subsequently fixed for all other temperatures and magnetic fields. Fits were performed over the range R=2.35R = 2.35–4.154.15 Å [green dashed curves in Figure 7(a) and Figure 8(a)]. Representative fits at 0 and 70 kOe are shown in Figure 7(a, Figure 7b) for 50 K and in Figure 8(a, Figure 8b) for 70 K (see Fig. S5 in [42] for the contributions from individual scattering paths). The magnetic-field dependence of the extracted parameters is summarized in Figure 7(c–Figure 7h) and Figure 8(c–Figure 8h). In all fits, S02S_0^2 was fixed at 0.793, and the resulting RR-factor (goodness of fit) lies in the range 0.004–0.006 at both temperatures.

Figure 7. (a, b) Fits to the ∣χ(R)∣|\chi(R)| and corresponding Im⁡[χ(R)]\operatorname{Im}[\chi(R)] spectra within the window indicated by the dotted green line [in panel (a)] at magnetic fields of 0 and 70 kOe, respectively, at 50 K. Each Im⁡[χ(R)]\operatorname{Im}[\chi(R)] spectrum is shifted downward for clarity. (c, d) Magnetic-field dependence of the average Eu–In1_1 and Eu–Eu scattering path distances, respectively. (e, f) Corresponding Debye-Waller factors. (g, h) Magnetic-field dependence of the Eu–In2_2 and Eu–In3_3 scattering path distances. Solid black lines indicate the field independent behavior of the various scattering distances and their corresponding Debye-Waller factors.

Figure 8. (a, b) Fits to the ∣χ(R)∣|\chi(R)| and corresponding Im⁡[χ(R)]\operatorname{Im}[\chi(R)] spectra within the window indicated by the dotted green line [in panel (a)] at magnetic fields of 0 and 70 kOe, respectively, at 70 K. Each Im⁡[χ(R)]\operatorname{Im}[\chi(R)] spectrum is shifted downward for clarity. (c, d) Magnetic-field dependence of the average Eu–In1_1 and Eu–Eu scattering path distances, respectively. (e, f) Corresponding Debye-Waller factors. (g, h) Magnetic-field dependence of the Eu–In2_2 and Eu–In3_3 scattering path distances. Solid black lines indicate the field independent behavior of the various scattering distances and their corresponding Debye-Waller factors.

Interestingly, the extracted Eu–In and Eu–Eu bond lengths, as well as their corresponding DW factors, exhibit no statistically significant field dependence up to 70 kOe at both 50 K (below TCT_{\mathrm{C}}) and 70 K (above TCT_{\mathrm{C}}). Since the DW factor captures contributions from both thermal and static disorder, the absence of any field-induced variation indicates that the applied magnetic field does not introduce additional local disorder, nor does it drive a measurable local distortion (e.g., magnetostriction) in the Eu coordination environment. While extremely subtle structural changes beyond the sensitivity of the present EXAFS measurements (≈0.01\approx0.01–0.020.02 Å) cannot be completely excluded, our results suggest that the second anomaly observed in the specific heat is unlikely to be associated with a magnetic-field-driven local structural modification in Eu2_2In.

Discussion

The change in the behavior of magnetocaloric properties of Eu2_2In, namely, deviation of the ΔSMmax⁡(H)\Delta S_{\mathrm{M}}^{\max}(H) curve, systematic variation of the local power exponent nn, and evolution of the UMC around 25 kOe, points toward a modification in the nature of the transition at this critical field. At the same time, the occurrence of an unphysically small value of the power exponent nn, the overshoot of the local exponent above n=2n = 2, and the persistence of dispersion in the UMC even for H=25H = 25–70 kOe indicate that the overall transition retains its first-order character. It is important to note that ΔSM\Delta S_{\mathrm{M}} is a cumulative quantity, and its values at higher fields necessarily incorporate contributions from the low-field region [see Eq. (1)]. As a result, even if the transition evolves toward second-order behavior for H>25H > 25 kOe, the imprint of the low-field first-order character would persist in the ΔSM\Delta S_{\mathrm{M}} curves at higher fields, giving rise to first-order-like features in all derived quantities.

To test this scenario, we apply the Banerjee criterion, which states that negative and positive slopes of Arrott plots (H/MH/M versus M2M^2) in the high-field region correspond to first-order and second-order transitions, respectively [54, 55]. Figure 9(a–Figure 9e) display the modified Arrott plots (MAPs) [M1/βM^{1/\beta} versus (H/M)1/γ(H/M)^{1/\gamma}] for Eu2_2In, constructed for different universality classes using their corresponding critical exponents (β\beta and γ\gamma) [56]. If the transition indeed crossed over to second-order at high fields, the MAPs would approach straight lines in the high-field regime for the correct choice of (β\beta, γ\gamma) [57, 58, 59]. In contrast, none of the tested universality classes yield linear behavior in the high-field region. The insets of Figure 9(a–Figure 9e), highlighting the 50–70 kOe field range, reveal a pronounced curvature that persists up to the highest fields investigated, confirming the robustness of the first-order character. Moreover, the slope of the MAPs in this regime changes from positive to negative with increasing temperature. To quantify this behavior, linear fits were performed in the 60–70 kOe range [see Figs. S6(a–e) of Ref. [42]], and the normalized slope is plotted in Figure 9(f). A distinct crossover from positive to negative slope occurs around 78 K for all universality classes, consistent with first-order-like behavior in which the critical exponents diverge at TcT_{\mathrm{c}}. These results unambiguously demonstrate that the first-order nature of the transition in Eu2_2In remains intact at least up to 70 kOe.

Figure 9. (a–e) Modified Arrott plots (MAPs) [M1/βM^{1/\beta} vs (H/M)1/γ(H/M)^{1/\gamma}] of Eu2_2In in the temperature range 40-90 K for different universality classes. Insets show an enlarged view of the high-field (50-70 kOe) region. (f) Normalized slopes of the MAPs obtained from 60-70 kOe field range for different universality classes.

The splitting of the specific-heat peak under an external magnetic field, together with its correlation with the transition temperature observed in magnetization, and the field-dependent variation in the magnitude of the two peaks, suggest that a field-induced subtle change in the crystal structure of Eu2_2In could be responsible for the LT peak in the CPC_{\mathrm{P}} curves. However, at the same time, magnetic-field-dependent EXAFS measurements rule out any detectable field-induced local structural changes in the sample within the experimental resolution. Recently, Hardy et al. proposed a mathematical model describing a “two-step” process in the first-order transformation of giant magnetocaloric materials [15]. In this framework, the transformation initiates and terminates through abrupt changes in the transformed fraction, accompanied by large exchanges of latent heat [15]. In the present case, the external field may induce a similar two-step-like process in Eu2_2In. Here it is important to mention that the field-induced splitting in the CP(T)C_{\mathrm{P}}(T) curves of Eu2_2In is fully reversible, where the doublet reverts back to a singlet upon removal of the external field. We note that different sample geometries were used in the present measurements, namely an irregular bulk piece for magnetization, a thin rectangular plate for specific-heat measurements, and a powdered sample for XAS. Consequently, demagnetization effects, which lead to differences between the applied and internal magnetic fields, may vary among the different measurement techniques, particularly in the low-field regime. However, this effect is not expected to significantly influence the overall field dependence or the main conclusions of this study. A related two-stage transition has also been reported in Nd2_2In, where the longitudinal strain first increases with magnetic field and then decreases at higher fields during its first-order transition at 108 K [22]. The authors correlate this two-stage transition in Nd2_2In with theoretical calculations for Eu2_2In, where the free energy with respect to FM order exhibits two minima [22, 31]. At this stage, we believe that more comprehensive investigations such as magnetic field dependent diffraction [7, 60] and/or a simultaneous probes of magnetic and structural responses [61, 62, 63] are required to clarify the origin of the two closely spaced peaks observed in the CP(T,H)C_P(T,H) curves of Eu2_2In.

Conclusion

In summary, we have investigated the temperature- and magnetic-field-dependent magnetization, specific heat, and local crystal structure of Eu2_2In across its first-order ferromagnetic–paramagnetic transition. The field dependence of ΔSM\Delta S_\mathrm{M} follows a power-law behavior for H≳25H \gtrsim25 kOe, but deviates at lower fields. The local field exponent nn increases near TcT_\mathrm{c} up to ∼\sim25 kOe and then decreases with further increase in field, which, together with the reduced dispersion in the universal master curve of ΔSM(H,T)\Delta S_\mathrm{M}(H,T) around 25 kOe, suggests a tendency of the transition to evolve toward second-order character above this critical field. Nevertheless, quantitative analyses of the magnetocaloric properties and modified Arrott plots unambiguously confirm that the overall first-order nature of the transition persists at least up to 70 kOe. While this behavior contrasts with theoretical predictions of a field-induced first- to second-order transition above ∼\sim25 kOe [32], it certainly indicates a subtle modification of the transition mechanism at higher fields. Specific-heat measurements reveal a clear field-induced splitting of the sharp, δ\delta-like anomaly into a doublet. A comparison between the CPC_P peak positions and the TcT_\mathrm{c} obtained from magnetization, together with the field-dependent changes in their relative amplitudes, suggests a nonmagnetic origin for the low-temperature peak. However, at the same time, EXAFS measurements show no detectable changes in the local crystal structure, ruling out a field-induced structural distortion in the sample within the experimental resolution. Taken together, these results point to a field-driven transformation of the transition in Eu2_2In from a one-step to a two-step process.

Acknowledgments

This work was performed at Ames National Laboratory and was supported by the Division of Materials Science and Engineering of the Office of Basic Energy Sciences, Office of Science of the U.S. Department of Energy (DOE). Ames National Laboratory is operated for the U.S. DOE by Iowa State University of Science and Technology under Contract No. DE-AC02-07CH11358. The XAS measurements were conducted at the Center for High-Energy X-ray Sciences (CHEXS), which is supported by the National Science Foundation (BIO, ENG and MPS Directorates) under award DMR-2342336. T. A. Tyson is supported by NSF Award DMR-2313456. Work at Argonne National Laboratory was supported by the U.S. DOE Office of Science-Office of Basic Energy Sciences, under Contract No. DE-AC0206CH11357.

APPENDIX

Additional magnetization data

Figure 10(a) shows the virgin magnetization isotherms of Eu2_2In measured from 0 to 70 kOe at different temperatures across TcT_\mathrm{c}. For T≤TcT \leq T_\mathrm{c}, the curves show the saturating behavior similar to that observed at 2 K [Figure 1(d)]; however for T>TcT > T_\mathrm{c}, the magnetic moment increases almost linearly for the lower applied magnetic fields and then show as step-like jump with further increase in the magnetic field [see Figure 10(a)]. The critical magnetic field (HcH_\mathrm{c}) for this jump in the magnetization, i.e., the magnetic field required to ferromagnetically align the spins above TcT_\mathrm{c}, increases with the temperature. In order to clearly present this, we show the normalized field derivative of the magnetization in Figure 10(b) for T>TcT > T_\mathrm{c}, where a monotonic shift in the peak position can be clearly observed to the higher field with increasing sample temperature. In Figure 10(c), we plot this peak position as a function of temperature, which shows that HcH_\mathrm{c} increases linearly with temperature for T>TcT > T_\mathrm{c}. Here, we perform the Gaussian fitting of the peaks in the vicinity of the transition to precisely estimate the peak position. The straight fit of the experimental data, as shown by the black solid line in Figure 10(c), gives a significant shift of 2.35 kOe/K.

The temperature dependent magnetization extracted from these virgin magnetization isotherms is presented in Figure 10(d) at the different magnetic fields. With increase in the magnetic field, the transition becomes broader with a monotonic shift in the transition temperature to the higher values. The normalized dM/dTdM/dT is plotted as a function of temperature in Figure 10(e) at different fields to show the field evolution of the TcT_\mathrm{c}. The field dependence of TcT_\mathrm{c} is shown in Figure 10(f), which shows the linear behavior up to 70 kOe and a straight fit of the data gives a shift of 0.37 K/kOe in the transition temperature. This linear dependence of the transition temperature of Eu2_2In on the external field makes it a suitable candidate for the device application in a wide range of the working temperature.

Additional specific heat data

Figure 11 shows the zero-field temperature dependent specific heat data of Eu2_2In (solid red circles). To account for the lattice (phonon) contribution to the specific heat of Eu2_2In, we measured the CP(T)C_P(T) data of its isostructural nonmagnetic reference compound Yb2_2In [18]. Since

Eu2In and Yb2In have different formula unit masses, a correction for the mass difference was considered. Within the Debye approximation, the phonon heat capacity depends on the reduced temperature T/θDT/\theta_D [see Eq. (5)]. The Debye temperature θD\theta_D depends on the formula unit mass MM as θD∝M−1/2\theta_D \propto M^{-1/2} [64]. Consequently, the measured temperature scale of Yb2In was rescaled according to the relation [64]

T∗=T(MYb2InMEu2In)1/2.(7)T^{*}=T\left(\frac{M_{\mathrm{Yb_2In}}}{M_{\mathrm{Eu_2In}}}\right)^{1/2}. \tag*{(7)}

The as-measured and mass-corrected CP(T)C_P(T) data of Yb2In are shown in Figure 11. No significant difference between the two datasets is observed in the present case. Therefore, the as-measured CP(T)C_P(T) data of Yb2In are used to account for the lattice heat capacity of Eu2In in all subsequent analyses [45, 65].

Figure 12 shows the magnetic-field dependence of the peak magnitudes of the two anomalies observed in the specific-heat data of Eu2In (see Figure 4), after deconvolution for H≥15H \ge15 kOe. The low-temperature (LT) peak is almost field invariant, whereas the high-temperature (HT) peak decreases monotonically with increasing field and approaches the LT value near H≈70H \approx70 kOe. The field invariance of the LT anomaly suggests a nonmagnetic (likely structural) origin of this feature.

Figure 13 shows the temperature dependence of the specific heat of Eu2In near the magnetic transition under different applied magnetic fields, measured using the long heat-pulse method in heating mode. The CP(T)C_P(T)

data of the isostructural nonmagnetic reference compound Yb2In, shown by the black solid curve, are used to estimate the lattice contribution to the specific heat of Eu2In. The magnetic entropy calculated by considering this lattice contribution is shown in the inset of Figure 13. It is observed that the magnetic entropy associated with the transition is nearly field independent and is significantly lower than the theoretical value expected for Eu2+^{2+}, as indicated by the horizontal dashed line in the inset of Figure 13. Notably, the zero-field magnetic entropy associated with the transition is close to that reported in

Ref. [19] (∼\sim5.7 J/Eu-mol-K), where a step-function approximation was used to estimate the lattice background in the transition region, despite the significantly larger peak value of CPC_P observed in Ref. [19]. This suggests that the transition in the present case is broader, possibly reflecting differences in compositional homogeneity between the two samples [66].

References

References

  1. [1]H. Hou, S. Qian, and I. Takeuchi, Materials, physics and systems for multicaloric cooling, Nat. Rev. Mater. 7, 633 (2022).
  2. [2]R. Kainuma, Y. Imano, W. Ito, Y. Sutou, M. Morito, S. Okamoto, O. Kitakami, and T. Kanomata, Magnetic-field-induced shape recovery by reverse phase transformation, Nature 439, 957 (2006).
  3. [3]J. Mira, F. Rivadulla, J. Rivas, A. Fondado, T. Guidi, R. Caciuffo, F. Carsughi, P. G. Radaelli, and J. B. Goodenough, Structural Transformation Induced by Magnetic Field and “Colossal-Like” Magnetoresistance Response above 313 K in MnAs, Phys. Rev. Lett. 90, 097203 (2003).
  4. [4]J. G. Roch, D. Miserev, G. Froehlicher, N. Leisgang, L. Spohnfeldner, K. Watanabe, T. Taniguchi, J. Klinovaja, D. Loss, and R. J. Warburton, First-Order Magnetic Phase Transition of Mobile Electrons in Monolayer MoS₂, Phys. Rev. Lett. 124, 187602 (2020).DOI
  5. [5]V. K. Pecharsky and K. A. Gschneidner, Jr., Giant Mag- netocaloric Effect in Gd₅(Si₂Ge₂), Phys. Rev. Lett. 78, 4144 (1997).
  6. [6]N. A. De Oliveira, P. J. Von Ranke, M. V. Tovar Costa, and A. Troper, Magnetocaloric effect in the intermetallic compounds RCo₂ (R = Dy, Ho , Er ), Phys. Rev. B 66, 094402 (2002).
  7. [7]V. K. Pecharsky, A. P. Holm, K. A. Gschneidner, and R. Rink, Massive Magnetic-Field-Induced Structural Transformation in Gd₅Ge₄ and the Nature of the Giant Magnetocaloric Effect, Phys. Rev. Lett. 91, 197204 (2003).
  8. [8]A. Kumar, A. Biswas, and Y. Mudryk, Stability of the first-order character of phase transition in HoCo₂, Journal of Applied Physics 136, 203902 (2024).DOI
  9. [9]W. Choe, V. K. Pecharsky, A. O. Pecharsky, K. A. Gschneidner, V. G. Young, and G. J. Miller, Making and Breaking Covalent Bonds across the Magnetic Transition in the Giant Magnetocaloric Material Gd₅(Si₂Ge₂), Phys. Rev. Lett. 84, 4617 (2000).
  10. [10]Y. Mozharivskyj, A. O. Pecharsky, V. K. Pecharsky, and G. J. Miller, On the high-temperature phase transition of Gd₅Si₂Ge₂, J. Am. Chem. Soc. 127, 317 (2005).
  11. [11]M. Piazzi, C. Bennati, and V. Basso, Thermodynamics of the heat-flux avalanches at the first-order magnetic transition in magnetocaloric materials, Phys. Rev. Appl. 8, 044023 (2017).DOI
  12. [12]F. Erbesdobler, C. R. H. Bahl, R. Bjørk, and K. K. Nielsen, Decided isofield calorimetry of La(Fe,Si,Mn)H reveals distributed magnetocaloric phase transitions, J. Appl. Phys. 127, 095102 (2020).DOI
  13. [13]C. Bennati, L. Gozzelino, E. S. Olivetti, and V. Basso, Heterogeneous nucleation and heat flux avalanches in La(Fe, Si)₁₃ magnetocaloric compounds near the critical point, Appl. Phys. Lett. 109, 231904 (2016).DOI
  14. [14]F. Guillou, H. Yibole, Z. Q. Ou, E. Brück, and O. Tegus, Large recalescence-like event at the first cooling across the magnetic transition of (Mn,Fe)₂(P,Si) magnetocaloric materials, Scr. Mater. 160, 81 (2019).DOI
  15. [15]V. Hardy, R. Hamane, F. Veillon, M. Risser, and F. Guillou, “Two-steps” process in the first-order transformation of giant magnetocaloric materials, Acta Mater. 231, 117869 (2022).DOI
  16. [16]P. J. Shamberger and F. S. Ohuhi, Hysteresis of the martensitic phase transition in magnetocaloric-effect Ni-Mn-Sn alloys, Phys. Rev. B 79, 144407 (2009).DOI
  17. [17]V. Provenzano, A. J. Shapiro, and R. D. Shull, Reduction of hysteresis losses in the magnetic refrigerant Gd₅Ge₂Si₂ by the addition of iron, Nature 429, 853 (2004).
  18. [18]A. Biswas, A. Kumar, P. Singh, and Y. Mudryk, From Complex Magnetic Ground States to Magnetocaloric Effects: A Review of Rare Earth R₂In Intermetallic Compounds, J. Phys. D: Appl. Phys. 59, 053001 (2026).
  19. [19]F. Guillou, A. K. Pathak, D. Paudyal, Y. Mudryk, F. Wilhelm, A. Rogalev, and V. K. Pecharsky, Non-hysteretic first-order phase transition with large latent heat and giant low-field magnetocaloric effect, Nat. Commun. 9, 2925 (2018).DOI
  20. [20]A. Biswas, N. A. Zarkevich, A. K. Pathak, O. Dolotko, I. Z. Hlova, A. V. Smirnov, Y. Mudryk, D. D. Johnson, and V. K. Pecharsky, First-order magnetic phase transition in Pr₂In with negligible thermomagnetic hysteresis, Phys. Rev. B 101, 224402 (2020).DOI
  21. [21]A. Biswas, R. K. Chouhan, A. Thayer, Y. Mudryk, I. Z. Hlova, O. Dolotko, and V. K. Pecharsky, Unusual first-order magnetic phase transition and large magnetocaloric effect in Nd₂In, Phys. Rev. Materials 6, 114406 (2022).DOI
  22. [22]W. Liu, F. Scheibel, T. Gottschall, E. Bykov, I. Dirba, K. Skokov, and O. Gutfleisch, Large magnetic entropy change in Nd₂In near the boiling temperature of natural gas, Appl. Phys. Lett. 119, 022408 (2021).DOI
  23. [23]B. Gegen, B. Huhe, Z.-Q. Ou, F. Guillou, and H. Yibole, Thermal Hysteresis and Reversibility of the Giant Magnetocaloric Effect at the Ferromagnetic Transition of Nd₂In, Materials 18, 3104 (2025).DOI
  24. [24]W. Cui, G. Yao, S. Qun, Q. Wang, J. Zhu, and S. Yang, Unconventional magnetocaloric phase transition in R₂In (R=Nd, Pr) with lambda-like specific heat and nonhysteresis, J. Mater. Sci. Technol. 101, 80 (2022).DOI
  25. [25]M. Forker, R. Mü̈ßeler, S. C. Bedi, M. Olzon-Dionysio, and S. D. De Souza, Magnetic and electric hyperfine interactions in the rare-earth indium compounds R₂In studied by ¹¹¹Cd perturbed angular correlations, Phys. Rev. B 71, 094404 (2005).DOI
  26. [26]H. Gamari-Seale, T. Anagnostopoulos, and J. K. Yakინthos, Magnetic characteristics of rare-earth indium R₂In (R =Y, Nd, Sm, Gd, Tb, Dy, Ho, Er, and Tm) intermetallic compounds, J. Appl. Phys. 50, 434 (1979).DOI
  27. [27]F. Guillou, H. Yibole, R. Hamane, V. Hardy, Y. B. Sun, J. J. Zhao, Y. Mudryk, and V. K. Pecharsky, Crystal structure and physical properties of Yb₂In and Eu₂₋ₓYbₓIn alloys, Phys. Rev. Mater. 4, 104402 (2020).
  28. [28]W. Zaela and A. Szytula, Crystal structure and magnetic properties of RE₂In compounds, J. Less-Common Met. 138, 123 (1988).
  29. [29]C. Ritter, D. H. Ryan, A. Provino, G. Lamura, Y. Mudryk, R. K. Chouhan, P. Singh, D. D. Johnson, V. K. Pecharsky, and P. Manfrinetti, Magnetic Ordering in Eu₂In and Eu₂Sn, Journal of Alloys and Compounds 980, 173573 (2024).
  30. [30]D. H. Ryan, D. Paudyal, F. Guillou, Y. Mudryk, A. K. Pathak, and V. K. Pecharsky, The first-order magnetoelectric transition in Eu₂In: A ¹⁵¹Eu Mössbauer study, AIP Advances 9, 125137 (2019).DOI
  31. [31]E. Mendive-Tapia, D. Paudyal, L. Petit, and J. B. Staunton, First-order ferromagnetic transitions of lanthanide local moments in divalent compounds: An itinerant electron positive feedback mechanism and Fermi surface topological change, Phys. Rev. B 101, 174437 (2020).DOI
  32. [32]B. P. Alho, P. O. Ribeiro, P. J. Von Ranke, F. Guillou, Y. Mudryk, and V. K. Pecharsky, Free-energy analysis of the nonhysteretic first-order phase transition of Eu₂In, Phys. Rev. B 102, 134425 (2020).
  33. [33]A. P. Holm, V. K. Pecharsky, K. A. Gschneidner Jr., R. Rink, and M. N. Jiramanus, X-ray powder diffractometer for in situ structural studies in magnetic fields from 0 to 35 K, Rev. Sci. Instrum. 75, 1081 (2004).DOI
  34. [34]V. Hardy, Y. Bréard, and C. Martin, Derivation of the heat capacity anomaly at a first-order transition by using a semi-adiabatic relaxation technique, J. Phys.: Condens. Matter 21, 075403 (2009).DOI
  35. [35]J. C. Lashley, M. F. Hundley, A. Migliori, J. L. Barrao, P. G. Pagliuso, T. W. Darling, M. Jaime, J. C. Cooley, W. L. Hults, L. Morales, D. J. Thoma, J. L. Smith, J. Boerio-G oates, B. F. Woodfield, G. R. Stewart, R. A. Fisher, N.E. Phillips, Critical examination of heat capacity measurements made on a quantum design physical property measurement system, Cryogenics 43, 369 (2003).DOI
  36. [36]G. Bunker, Introduction to XAFS: A Practical Guide to X-Ray Absorption Fine Structure Spectroscopy (Cambridge University Press, Cambridge, 2010).
  37. [37]A. Kumar, R. Shukla, R. Kumar, R. J. Choudhary, S. N. Jha, and R. S. Dhaka, Probing the electronic and local structure of Sr₂₋ₓLaₓCoNbO₆ using near-edge and extended X-ray absorption fine structures, Phys. Rev. B 105, 245155 (2022).DOI
  38. [38]B. Ravel and M. Newville, ATHENA, ARTEMIS, HEP-HAESTUS: Data analysis for x-ray absorption spectroscopy using IFEFFIT, J. Synchrotron Rad. 12, 537 (2005).
  39. [39]M. H. Phan and S. C. Yu, Review of magnetocaloric effect in manganite materials, J. Magn. Magn. Mater. 308, 325 (2007).
  40. [40]A. Kumar and R. S. Dhaka, Unraveling magnetic interactions and the spin state in insulating Sr₂₋ₓLaₓCoNbO₆, Phys. Rev. B 101, 094434 (2020).
  41. [41]V. Franco, J. S. Blázquez, and A. Conde, Field dependence of the magnetocaloric effect in materials with a second order phase transition: A master curve for the magnetic entropy change, Applied Physics Letters 89, 222512 (2006).DOI
  42. [42]A. Kumar, A. Biswas, T. A. Tyson, D. Haskel, C. J. Pollock, and Y. Mudryk, See the supplementary information for the details of the magnetization, specific heat, and XAS data.
  43. [43]J. Y. Law, V. Franco, L. M. Moreno-Ramírez, A. Conde, D. Y. Karpenkov, I. Radulov, K. P. Skokov, and O. Gutfleisch, A quantitative criterion for determining the order of magnetic phase transitions using the magnetocaloric effect, Nat. Commun. 9, 2680 (2018).DOI
  44. [44]C. M. Bonilla, J. Herrero-Albillos, F. Bartolomé, L. M. García, M. Parra-Borderías, and V. Franco, Universal behavior for magnetic entropy change in magnetocaloric materials: An analysis on the nature of phase transitions, Phys. Rev. B 81, 224424 (2010).
  45. [45]V. K. Anand, D. T. Adroja, and A. D. Hillier, Ferromagnetic cluster spin-glass behavior in PrRhSn3, Phys. Rev. B 85, 014418 (2012).
  46. [46]C. Kittel, Introduction to Solid State Physics, 8th ed. (Wiley, New York, 2005).
  47. [47]A. Kumar, A. Biswas, and Y. Mudryk, Magnetic glassiness in noncentrosymmetric Sm7Pd3: Interplay of magnetic frustration, long-range order, and frozen domains, Phys. Rev. B 111, 174448 (2025).
  48. [48]W. Liu, F. Scheibel, N. Fortunato, I. Dirba, T. Gottschall, H. Zhang, K. Skokov, and O. Gutfleisch, Role of Debye temperature in achieving large adiabatic temperature changes at cryogenic temperatures: A case study on Pr2In, Phys. Rev. B 109, L140407 (2024).DOI
  49. [49]V. K. Pecharsky, K. A. Gschneidner Jr., Y. Mudryk, and D. Paudyal, Making the most of the magnetic and lattice entropy changes, J. Magn. Magn. Mater. 321, 3541 (2009).
  50. [50]A. Kumar, P. Singh, A. Doyle, D. L. Schlagel, and Y. Mudryk, Multiple magnetic interactions and large inverse magnetocaloric effect in TbSi and TbSi0.6Ge0.4, Phys. Rev. B 109, 214410 (2024).DOI
  51. [51]W. B. Jiang, M. Smidman, W. Xie, J. Y. Liu, J. M. Lee, J. M. Chen, S. C. Ho, H. Ishii, K. D. Tsuei, C. Y. Guo, Y. J. Zhang, Hanoh Lee, and H. Q. Yuan, Antiferromagnetism with divalent Eu in EuNi5As3, Phys. Rev. B 95, 024416 (2017).DOI
  52. [52]N. M. Souza-Neto, J. Zhao, E. E. Alp, G. Shen, S. V. Sinogeikin, G. Lapertot, and D. Haskel, Reentrant Valence Transition in EuO at High Pressures: Beyond the Bond-Valence Model, Phys. Rev. Lett. 109, 026403 (2012).
  53. [53]X. Tan, G. Fabbris, D. Haskel, A. A. Yaroslavtsev, H. Cao, C. M. Thompson, K. Kovnir, A. P. Menushenkov, R. V. Chernikov, V. O. Garlea, and M. Shatruk, A transition from localized to strongly correlated electron behavior and mixed valence driven by physical or chemical pressure in ACdO2As2 (A = Eu and Ca), J. Am. Chem. Soc. 138, 2724 (2016).DOI
  54. [54]A. Arrott and J. Noakes, Approximate Equation of State for Nickel Near its Critical Temperature, Phys. Rev. Lett. 19, 786 (1967).DOI
  55. [55]S. K. Banerjee, On a generalised approach to first and second order magnetic transitions, Phys. Lett. 12, 16 (1964).DOI
  56. [56]J. C. Le Guillou and J. Zinn-Justin, Critical exponents from field theory, Phys. Rev. B 21, 3976 (1980).
  57. [57]P. Sarkar, P. Mandal, A. K. Bera, S. M. Yusuf, L. S. Sharath Chandra, and V. Ganesan, Field-induced first-order to second-order magnetic phase transition in Sm0.52Sr0.48MnO3, Phys. Rev. B 78, 012415 (2008).DOI
  58. [58]S. Rößler, U. K. Rößler, K. Nenkov, D. Eckert, S. M. Yusuf, K. Dörr, and K.-H. Müller, Rounding of a first-order magnetic phase transition in Ga-doped La0.67Ca0.33MnO3, Phys. Rev. B 70, 104417 (2004).
  59. [59]A. Kumar, A. Biswas, and Y. Mudryk, Nontrivial critical behavior at magnetic transitions: A case study of Sm7Pd3, Phys. Rev. B 112, 224416 (2025).
  60. [60]Y. Mudryk, D. Paudyal, V. K. Pecharsky, K. A. Gschneidner, S. Misra, and G. J. Miller, Controlling Magnetism of a Complex Metallic System Using Atomic Individualism, Phys. Rev. Lett. 105, 066401 (2010).
  61. [61]A. Aubert, K. Skokov, G. Gomez, A. Chirkova, I. Radulov, F. Wilhelm, A. Rogalev, H. Wende, O. Gutfleisch, and K. Ollffs, Simultaneous multi-property probing during magneto-structural phase transitions: An element-specific and macroscopic hysteresis characterization at ID12 of the ESRF, IEEE Transactions on Instrumentation and Measurement 71, 1 (2022).DOI
  62. [62]K. P. Skokov, I. A. Y. Karpenkov, D. Y. Karpenkov, I. A. Radulov, D. Gunzing, B. Eggert, A. Rogalev, F. Wilhelm, J. Liu, Y. Shao, K. Ollffs, M. E. Gruner, H. Wende, and O. Gutfleisch, A multi-stage, first-order phase transition in LaFe11.8Si1.2: Interplay between the structural, magnetic, and electronic degrees of freedom, Applied Physics Reviews 10, 031408 (2023).
  63. [63]D. Yu. Karpov, A. Yu. Karpenkov, K. P. Skokov, I. A. Radulov, M. Zheleznyi, T. Faske, and O. Gutfleisch, Pressure Dependence of Magnetic Properties in La(Fe,Si)13: Multistimulus Responsiveness of Caloric Effects by Modeling and Experiment, Phys. Rev. Applied 13, 034014 (2020).
  64. [64]V. K. Anand and D. C. Johnston, Antiferromagnetism in EuCu2As2 and EuCu1.82Sb2 single crystals, Phys. Rev. B 91, 184403 (2015).DOI
  65. [65]V. K. Anand, Z. Hossain, G. Chen, M. Nicklas, and C. Geibel, Heavy fermion behavior in PrRh2B2C: Excitonic mass enhancement, Phys. Rev. B 79, 113107 (2009).DOI
  66. [66]N. J. Jones, H. Ucar, J. J. Ipus, M. E. McHenry, and D. E. Laughlin, The effect of distributed exchange parameters on magnetocaloric refrigeration capacity in amorphous and nanocomposite materials, J. Appl. Phys. 111, 07A334 (2012).

Paper details

vlnc:W_ce27pk72
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