Enhancement in magnetocaloric properties of holmium chromite by gadolinium substitution Free

In the past decade, magnetoelectric multiferroics (ME MFs) with both ferroelectric and magnetic orders have been considered attractive for their potential applications in devices such as random access memory, magnetic field sensors, and capacitors.^1–5 A few rare earth manganites (RMnO₃) have been regarded as single phase ME MFs with strong coupling between ferroelectric and magnetic orders;^1,6,7 however, they become MF only at relatively low temperatures (e.g., ∼27 K for TbMnO₃).⁷ Recently, great attention has been paid to rare-earth chromites (RCrO₃) since their magnetically driven ferroelectricity temperatures are much higher than their manganite counterparts.^8–12 These RCrO₃ materials have also been intensively investigated due to their magnetocaloric effect (MCE), which is the basis of magnetic refrigeration, a prospective technology to replace conventional vapor-compression refrigeration that contributes to a global energy problem.^13–15 For example, bulk DyCrO₃ was reported to have a large MCE value at ∼8.4 J/kg K and relative cooling power (RCP) 217 J/kg at ∼20 K and at 4 T, which renders it useful for refrigeration application below 30 K.¹⁰ In order to further improve the MCE property of DyCrO₃, substitution at the A-site (by another rare-earth element) or at the B-site (Mn, Fe, etc.) has been considered.^16–18 For example, B-site substituted DyCrO₃, i.e., bulk powder DyCr_0.5Fe_0.5O₃ sample, was found to have a larger MCE value ∼10.5 J/kg at 5 K and 4 T than that of pure DyCrO₃ powder sample, meanwhile the Cr³⁺ ordering temperature increased (261 K) compared to that of pure DyCrO₃ (146 K).¹⁷ On the other hand, A-site substitution in DyCrO₃ has resulted in mixed MCE values. For example, the A-site substituted DyCrO₃, i.e., bulk Dy_0.7Y_0.3CrO₃ or Dy_0.7Ho_0.3CrO₃ showed slightly lower MCE values than pure DyCrO₃, while Dy_0.7Er_0.3CrO₃ showed larger MCE value (10.92 J/kg K) than DyCrO₃ or even DyCr_0.5Fe_0.5O₃.¹⁶ 

Another rare earth chromite, HoCrO₃, was also reported to have a large MCE value ∼7.2 J/kg K and RCP ∼408 J/kg at 20 K and at a magnetic field of 7 T.¹⁹ B-site substitution in HoCrO₃ has also been considered to further explore the resulting MCE properties. Fe substitution at the B-site in HoCrO₃ (i.e., bulk HoCr_0.7Fe_0.3O₃) has, however, resulted in smaller MCE (∼6.8 J/kg K) and RCP (∼387 J/kg) values at 20 K and at a magnetic field of 7 T.¹⁹ On the other hand, the effect of A-site substitution on the MCE properties of HoCrO₃ has not been explored yet. So, it is worthwhile to investigate how the MCE properties of HoCrO₃ are altered by substitution at the A-site. To select a suitable rare earth element as an A-site dopant, we note that among all rare earth elements, Gd based materials have been investigated widely and they also exhibit excellent MCE properties. For example, Gd single crystal was shown to have MCE value of ∼8.9 J/(kg K) and RCP value of ∼461.6 J/kg at room temperature and at a field of 5 T, and thus it is considered useful for magnetic refrigeration near room temperature.^20–22 Also, Gd alloys were reported to have superior MCE property.^13,22,23 A large MCE value was reported for Gd₅(Si₂Ge₂) with an improvement of MCE value by a factor of 2 compared to Gd single crystal, which is attributed to arise due to a first-order phase transition at 276 K and unique magnetization dependence on field.²⁰ Recently, Ren et al.²⁴ showed that (Gd, In) solid solution has slightly smaller MCE values than Gd single crystal, but the transition temperature is lower and tunable making the solid solution suitable for magnetic refrigeration at lower temperatures. Moreover, the MCE properties of Gd based complex oxides have also been studied. Yu et al. reported GdMnO₃ single crystal shows the MCE value of 31 J/(kg K) at 19 K and at a magnetic field of 7 T along the c-axis.²⁵ In another recent study, Yin et al. reported giant MCE value at ∼31.6 J/kg K for a field of 4.4 T in single crystal GdCrO₃ along c-axis. However, the usage of single crystals for practical magnetic refrigeration is envisioned to be limited owing to the direction dependent MCE properties along with the difficulty in synthesizing single crystals. Guided by the observation of superior MCE properties in Gd containing single crystalline materials in the aforementioned studies, we believe it would be worthwhile exploring the effect of Gd substitution (A-site dopant) on the resulting MCE properties in bulk HoCrO₃.

In the current work, we employed a citrate route to prepare HoCrO₃, Ho_0.67Gd_0.33CrO₃, and GdCrO₃ bulk powder samples. Their structural, magnetic, and magnetocaloric properties have been investigated in detail and are presented here. MCE value of the polycrystalline GdCrO₃ bulk powder sample was found to be higher than that of all polycrystalline RMnO₃ and RCrO₃ bulk powder materials.

In order to make phase-pure HoCrO₃ (HCO), Ho_0.67Gd_0.33CrO₃ (HGCO), and GdCrO₃ (GCO) bulk powder samples, first the stoichiometric amounts of high-purity Ho(NO₃)₃, Gd(NO₃)₃, and Cr(NO₃)₃ precursors were separately dissolved in water and then mixed together. Citric acid was then added to this mixture followed by drying on a hot plate at 100 °C. The resultant dry powder was annealed at 900 °C in oxygen atmosphere for 2 h. The crystal structure and phase purity of the samples were characterized by x-ray diffraction (XRD) (Bruker D2 x-ray diffractometer using Cu-Kα radiation λ = 1.542 Å). The structural parameters (lattice constants, bond angle/length, etc.) of the samples were obtained by Rietveld refinement of the experimental XRD data performed using Fullprof Suite software. The room-temperature Raman spectroscopic measurements were performed by Renishaw System 2000 using a 514 nm-wavelength Ar-ion laser. The dc magnetic property was measured using the Vibrating Sample Magnetometer (VSM) attached to the Evercool physical property measurement system (PPMS) manufactured by Quantum Design. The temperature dependent magnetization was measured from 5 K to 300 K with an applied field of 50 Oe. The field dependent magnetization measurement was performed with the applied field up to 7 T from 5 K to 145 K (5 K step). The ac susceptibility was recorded using the ac susceptibility option attached to the PPMS from 5 K to 300 K under an applied ac magnetic field (amplitude ∼10 Oe, frequencies of 500 and 5000 Hz).

The experimental XRD patterns obtained for HCO, HGCO, and GCO bulk powders along with the Reitveld refined patterns are shown in Figures 1(a)–1(c). For each sample, the experimental XRD pattern could be refined with an orthorhombically distorted perovskite structure (space group Pbnm). No extra peaks and the goodness of the Reitveld refinements suggests that the samples were phase pure. Various structural parameters, such as the lattice constants, Cr-O-Cr bond angles, and strain parameters, as obtained from the Reitveld refinements, are summarized in Table I. The lattice parameters of HCO and GCO agree well with the previous reports^26,27 and those of HGCO are intermediate between the end compositions (HCO and GCO). The orthorhombic distortions of the unit cell from the ideal cubic structure, defined by the orthorhombic strain factor S = 2(b−a)/(b+a),²⁸ were calculated and summarized in Table I as well. Table I also lists the average ionic radii (R_a,avg) of the A-site ion, where R_a,avg of HGCO is calculated by $0.33*rGd2+0.67*rHo2$⁠, considering the atomic number ratio and ionic radii of Ho³⁺ and Gd³⁺ ions. As can be seen in Table I, the lattice parameters and hence the lattice volume increase with R_a,avg, i.e., with increasing Gd substitution at the Ho-site. The sizable mismatch between R–O and Cr–O bond lengths in the Goldschmidt tolerance factor {t ≡ (R–O)/[√2*(Cr–O)]} of < 1 results in significant cooperative rotations of the CrO₆ octahedra, which decreases with increasing R_a,avg. This leads to the bending of the Cr–O–Cr angle and hence distortion. It should be noted that with the addition of Gd at the Ho-site in HCO, bond angles (in-plane Cr1–O1–Cr1 and out-of-plane Cr1–O2–Cr1) increase and the strain factor (S) decreases, indicative of the less distortion in the structure when higher ionic radius Gd-ion is substituted in place of the smaller ionic radius Ho-ion. This behavior is similar to that observed in the rare earth manganite (RMnO₃) counterparts.²⁹ However, the main difference between RMnO₃ and RCrO₃ compounds is that RCrO₃ compounds form orthorhombically distorted perovskite structure for all R, whereas RMnO₃ compounds have that structure only for R = La-Dy.³⁰ 

Raman spectroscopy is also very useful to study the phase purity and crystal structural distortion. Previous study indicates that orthorhombic Pbnm perovskite structure has 24 active Raman modes: 7A_g+5B_1g+7B_2g+5B_3g, 12 of which occurs in the 100–600 cm⁻¹ range.³¹ Figure 2 shows the Raman spectra of HCO, HGCO, and GCO bulk powders, where all the observed peaks can be assigned to Raman phonon modes according to the report of Weber et al.,³¹ further confirming that the bulk samples in the present study are phase pure. The Raman modes, such as B_2g(1), B_2g(2), A_g(4), A_g(6), show the systematic shift towards lower wave number with increasing Gd substitution in HCO. This result is consistent with the report of Weber et al., which presented a systematic shift in the position of these modes with decreasing A-site ionic radius. With decreasing average ionic radius at the A-site, the modes below 200 cm⁻¹ not only shift toward the lower wave numbers but also become more intense. Moreover, the two groups of adjacent modes—(B_1g(1) and A_g(3), B_1g(2) and A_g(5)) of HCO sample—are observed slightly closer in HGCO sample and finally form an indistinguishable broad peak in Raman spectrum of GCO sample.³¹ Since both the XRD patterns and the Raman spectra show a systematic shift in the positions of the observed peaks without noticeable broadening, it is clear that the present synthesis method was successful in substituting Gd for Ho in HCO without phase separation or impurity phases.

To study the effect of Gd substitution on magnetic properties of HCO, the temperature-dependent dc magnetization data in both zero field-cooled (ZFC) and field cooled (FC) conditions were measured with a magnetic field of 50 Oe as presented in Figure 3. For all samples, divergence of the ZFC and FC modes was observed below a certain temperature (Néel temperature, T_N), which is caused by a weak G-type antiferromagnetic ordering of the Cr³⁺ ions. By plotting the derivative of FC magnetization data (dM/dT) as a function of temperature, as shown in the insets of Figure 3, the T_N (where Cr³⁺ orders) values were determined to be 140, 148, and 167 K for HCO, HGCO, and GCO samples, respectively. The T_N values for HCO and GCO are in good agreement with previous reports on bulk samples in literatures.^9,32 The maximum value of magnetization was found to be 30.8, 18.8, and 1.5 emu/g for bulk HCO, HGCO, and GCO, respectively. Magnetic moments of Ho³⁺ and Gd³⁺ ions are 10.6 μ_B and 7.94 μ_B, respectively.³³ The higher magnetic moment of the Ho³⁺ ions along with cooperative alignment between Ho³⁺ moments and the net Cr³⁺ moments at applied field of 50 Oe leads to higher magnetization in HCO sample. On the other hand, lower magnetic moment of Gd³⁺ ion, which anti-aligns with the net Cr³⁺ moments,³⁴ leads to the lowering of magnetization values in Gd based samples as seen in Figures 3(b) and 3(c). We also note that the T_N and maximum magnetization values for HGCO sample lie between those of the two parent compounds, HCO and GCO, which suggests that these values could be made tunable by controlling the Ho³⁺/Gd³⁺ ratio in the Ho_1−xGd_xCrO₃ solid solution.

In RMnO₃, the Mn-O-Mn bond angles have been shown to have tremendous impact on their structure and magnetic properties.^6,35 In RCrO₃ system studied here, the Cr-O-Cr bond angle and R_a,avg (from Table I) were also found to directly influence the T_N. As presented in Figure 4, the in-plane Cr-O-Cr bond angle and R_a,avg of RCrO₃ (R = Ho, Gd, Nd, and La, where the data of NdCrO₃ and LaCrO₃ are extracted from Reference 36) have almost linear relationship with the experimentally measured T_N of the samples. Therefore, the effect of Gd substitution on the magnetic property of HCO is directly correlated with the modifications in its crystal structure.

Applied dc magnetic field dependence of the magnetization data for both ZFC and FC modes as a function of temperature is shown in Fig. 5 for GCO. The ZFC magnetization data in Fig. 5(a) maximize at 7 K for the data with applied field of 50–300 Oe when the temperature is decreased from 300 K and then it decreases with further lowering of temperature. This maximum at 7 K is interpreted by the spin reorientation of Cr³⁺ and Gd³⁺ ions from high-temperature Γ₄(G_x, A_y, F_z) phase to low-temperature Γ₂(F_x, C_y, G_z) phase.³⁷ Moreover, the spin reorientation behavior is applied field-dependent and only exists at field 50–300 Oe, whereas at field ≥500 Oe, the spin reorientation behavior was found to be suppressed, because the rotation of the magnetic moments or magnetic domain becomes more oriented at higher fields.³⁸ Figure 5(b) shows that on lowering the temperature from 300 K, the FC magnetization data with an applied field of 50 Oe first decrease to zero at ∼125 K and then the magnetization value becomes negative. Upon further lowering of temperature, the magnetization value becomes positive again at ∼10 K and reaches maximum value of ∼1.5 emu/g at 5 K. This phenomenon is called temperature-induced magnetization reversal (TMR)³² that arises from the competition of the two magnetic subsystems (Cr³⁺ ions and Gd³⁺ ions) whose magnetic ordering is in anti-parallel directions. However, it should be noted that the TMR behavior does not exist in pure HCO and in Gd substituted HCO (i.e., HGCO), this is because of the larger magnetic moment and mass ratio of the Ho³⁺ ions as compared to that of the Gd³⁺ ions. Note that in GCO single crystal, the TMR behavior was found to be dependent on the applied magnetic field and it disappeared with field higher than ∼300 Oe applied along c-axis.³² In the present work, the TMR behavior of GCO bulk (polycrystalline) sample under different magnetic fields (50–1500 Oe) was measured and is presented in Figure 5(b). With increasing applied magnetic field from 50 to 1500 Oe, the magnetization value at low temperature increases systematically. To describe TMR behavior, the two temperatures, where the magnetization reaches zero (lower one defined as T₀ and higher one as T₀′ as marked for 50 Oe data in the Figure 5(b)), are plotted in the inset of Figure 5(b). It can be seen that T₀ and T₀′ change linearly with the applied field from 50 to 1000 Oe. At applied field of 1500 Oe, the TMR behavior disappears because of stronger remnant magnetization in the sample, which could not be reversed by the temperature change. This critical field of 1500 Oe observed in present polycrystalline bulk GCO sample is higher than 300 Oe reported for the GCO single crystal. This could be because in GCO bulk sample, the magnetization is isotropic and the net magnetization is smaller than that in single crystal at the same field, and thus, higher magnetic field is required to reverse the direction of magnetization. It is noted that the TMR and spin reorientation behavior were not observed in HGCO sample because of large mass ratio of Ho³⁺.

Above T_N, the samples show paramagnetic behavior and FC susceptibility (χ) can be fitted by the Curie-Weiss law: χ = C/(T-θ), where magnetic susceptibility χ = M/(m*H), (M—magnetic moment, m— the mass of the sample, H—the applied magnetic field), C is Curie constant, and θ is the Weiss temperature. The fitting results are summarized in Table II. The effective magnetic moment (μ_eff) was calculated by C values using $μeff=3kBCN$⁠, where k_B is the Boltzmann constant and N is the Avogadro constant.³³ The values of μ_eff for different samples were also compared with the theoretical values calculated using the free ionic moments: $μ′eff=μCr2+(1−x)μHo2+xμGd2$⁠, where μ_Cr, μ_Ho, and μ_Gd are the free ionic moments of Cr³⁺ (3.8 μ_B), Ho³⁺ (10.4 μ_B), and Gd³⁺ (8.9 μ_B), respectively, and x is the Gd substitution fraction [Ho_1−xGd_xCrO₃]. The calculated values of μ_eff and μ′_eff for all the samples are also included in Table II. Close agreement between the μ_eff and μ′_eff values for each composition suggests that the present samples are phase-pure and in the correct substitution ratio. The Weiss temperature, θ, for all the present samples are negative, which is indicative of the antiferromagnetic behavior for all the samples studied here. The value of θ of the present GCO sample (−71.9 K) is smaller than what has been reported earlier for GCO bulk sample (−35 K).³⁸ 

Complementary to the dc magnetic measurement, the ac magnetization data can provide additional magnetic properties.^19,39 In Figure 6, the temperature dependent ac magnetization data (real part χ′, imaginary part χ″) of the HCO, HGCO, and GCO samples under an applied ac magnetic field (amplitude ∼10 Oe, frequencies ∼500 and 5000 Hz) are presented. The χ′(T) and χ″(T) data also revealed that the Néel temperature, T_N, of HCO and HGCO were 140 and 148 K, respectively, which were in good agreement with the dc magnetic results mentioned above. For HCO and HGCO bulk samples, an additional anomaly at ∼8 K was observed due to the ordering of rare-earth ions Ho³⁺, which was not revealed in dc magnetic measurement presented above.

To further investigate the magnetic properties, the field-dependent magnetization at 5, 50, 100, 145 K of the present samples was measured and is presented in Figure 7. Below T_N, all the present samples show antiferromagnetic behavior due to the canted antiferromagnetic ordering of Cr³⁺ ions. Comparatively, the samples are paramagnetic above T_N because of the linear paramagnetic behavior of Cr³⁺ and Ho³⁺/Gd³⁺ ions. At 5 K and 40 kOe, the magnetization values of HCO, HGCO, and GCO are 85.3, 96.3, and 100.3 emu/g, compared to the respective values of 40.1, 34.4, and 21.6 emu/g at 50 K and 40 kOe. It can be seen that the magnetization value of GCO is largest at 5 K but smallest at 50 K of the three samples, which can be interpreted by the spin reorientation at ∼8 K in the GCO sample (Fig. 3(c)). The magnetization values of HGCO are intermediate between those of HCO and GCO because of Gd substitution. Furthermore, HCO shows bigger magnetic hysteresis than GCO, which is in good agreement with literature.^9,32 This is probably because Gd³⁺ moments align antiparallel with Cr³⁺ moments and Ho³⁺ moments aligns with Cr³⁺ moments near the rare-earth ordering temperature.³⁴ To characterize the magnetic hysteresis, the temperature dependence of coercive field (H_c) and remnant magnetization (M_R) of the samples were extracted from the magnetic hysteresis data and are shown in Figures 8(a) and 8(b), respectively. Of immediate note is the qualitative similarity between HCO and HGCO data, which probably arises from the dominant contribution of heavier Ho ions present in these compositions. For HCO and HGCO, H_c shows a plateau ∼2600 Oe in the temperature range 5–100 K and then quickly decreases to zero near T_N. On the other hand, the coercive field, H_c, for GCO increases with increasing temperature and reaches a maximum (∼1300 Oe) at 40 K followed by gradual reduction to zero near T_N. Also, note that for GCO, H_c values are much smaller than those of HCO (or HGCO) in the entire temperature range studied may be due to smaller magnetic moment of Gd-ion (compared to Ho ions). At lower temperatures, the weaker R³⁺–Cr³⁺ exchange coupling becomes operative due to the lowering of thermal energy. Thus, the magnetic moment of the Gd-ion can now rotate with the Cr³⁺ moment, thereby lowering the measured H_C. This effect becomes increasingly important with decrease in temperature, providing a qualitative interpretation of the data in Fig. 8(a). The R³⁺-R³⁺ interaction in these compounds is weak and is not expected to significantly affect the magnetic properties above the rare-earth ordering temperature, leaving only the anisotropic component of R³⁺-Cr³⁺ interaction to account for the observed differences in the H_c values of the present samples. In addition, the M_R values of all three samples decrease monotonically with increasing temperature from a maximum of 33.5 emu/g (for HCO), 20.1 emu/g (for HGCO), and 7.4 emu/g (for GCO) at 5 K to zero at ∼150 K. The M_R values for HGCO lie in between those for HCO and GCO at all temperatures. Both H_c and M_R values become nearly zero for all three samples near their respective T_N, further illustrating their paramagnetic behavior above this temperature.

The MCE properties of the present samples were evaluated by calculating two factors: (i) magnetic entropy change ΔS_M(T,H) given by¹³ 

and (ii) relative cooling power (RCP) calculated by

where T₁ and T₂ are the high and low temperature limits in the refrigeration cycle.¹⁶ Figure 9 shows the temperature dependent ΔS_M(T,H) values for the present samples at different applied magnetic fields (1–7 T) derived from Eq. (1). The value of ΔS_M(T,H) for HCO enhances with increasing temperature and reaches a maximum at ∼20 K and then decreases with increasing temperature, whereas the value of ΔS_M(T,H) for GCO maximizes at 5 K and then decreases with increasing temperature. Note that ΔS_M(T,H) values of HGCO lie in between that of HCO and GCO not only in magnitudes but also in the temperature dependent behavior. At 1 T and 2 T, ΔS_M(T,H) values for HGCO shows a maximum at ∼20 K, similar to the trend for HCO. On the contrary, ΔS_M(T,H) values for HGCO decrease monotonically with increasing temperature for applied fields above 2 T, similar to the trend for GCO. At an applied field of 7 T, the maximum value of ΔS_M(T,H) for HCO, HGCO, and GCO is 7.2 J/kg K at 20 K, 14.0 J/kg K at 5 K, and 31.6 J/kg K at 5 K, respectively. Therefore, Gd substitution at the A-site of HCO was found to considerably improve the MCE properties of pure HCO, and GCO showed very large MCE values. Several factors might contribute towards the beneficial effects of Gd substitution in improving the MCE properties of HCO. First, the MCE value (ΔS_M(T,H)) is proportional to the derivative of magnetization versus temperature (dM/dT) according to Eq. (1). The dM/dT value of HCO is smaller because M changes monotonically with temperature, whereas the spin-reorientation in GCO results in large dM/dT value and thus larger MCE value (see Figures 3(a)–3(c)) is observed. Second, HCO sample exhibits much larger magnetic hysteresis compared to the GCO sample (see Fig. 7) and thus more energy is lost in the thermal process, resulting in smaller MCE values for HCO as compared to GCO. This agrees well with the report of Phan and Yu¹³ that materials with zero magnetic hysteresis are better suited for magnetic refrigeration. Third, it has been noted that the MCE value tends to reach maximum at 2 K in GCO single crystal, which is close to the ordering temperature of Gd³⁺ (2.3 K).³² Thus, the large MCE in GCO and HGCO may have contributions arising from the Gd³⁺-Gd³⁺ magnetic interactions.

The MCE values of HCO, HGCO, and GCO bulk samples are summarized and compared with earlier reports in Table III. This MCE value of the present GCO bulk is higher than that of all RMnO₃ and RCrO₃ bulk (polycrystalline) materials. However, the value of ΔS_M(T,H) for GCO bulk powder (∼19.0 J/kg K at 4 T) is lower than that of the GCO single crystal, which was reported to be ∼26.1 J/kg K at a field of ∼4 T.³² As the MCE property is direction-dependent for single crystals, smaller MCE values for bulk powder samples (compared to single crystals) are expected because of averaging effect.⁴⁰ 

Figure 10 shows the figure of merit for magnetic refrigeration, the RCP values of the samples (T₁ was chosen as 150 K for calculation using Eq. (2)). At 5 K and an applied field of ∼7 T, the RCP values of HCO, HGCO, and GCO bulk samples are 408, 490, and 546 J/kg, respectively (also summarized in Table III). Again, the RCP value of HCO improves considerably by Gd substitution at the A-site. ΔS_M and RCP values of the present GCO bulk sample were the highest among the reported values rendering it to be useful for potential magnetic refrigeration applications.

In conclusion, HoCrO₃, Ho_0.67Gd_0.33CrO₃, and GdCrO₃ powder samples were synthesized via a solution route. X-ray diffraction data revealed that the crystal structure of each composition was orthorhombically distorted perovskite (space group Pbnm) and no impurity phase was observed. The dc magnetic measurement showed that the ordering temperature of Cr³⁺ ions (T_N) is ∼140 K, 148 K, and 167 K for HoCrO₃, Ho_0.67Gd_0.33CrO₃, and GdCrO₃ samples, respectively, indicating that Gd substitution increased T_N of HoCrO₃. The ac magnetic measurements confirmed the Cr³⁺ ordering temperatures of HoCrO₃ and Ho_0.67Gd_0.33CrO₃ and also revealed ordering temperature of rare-earth ions at ∼8 K in HoCrO₃ and Ho_0.67Gd_0.33CrO₃. The isothermal magnetization data indicated that the magnetic behavior of the present samples changed from low-temperature (below T_N) canted antiferromagnetic to high-temperature (above T_N) paramagnetic. These features were characterized by the temperature dependent coercive field and remnant magnetization behavior. At an applied field of 7 T, the maximum magnetocaloric values were 7.2, 14.0, and 31.6 J/kg K, and the relative cooling power were 408, 490, and 546 J/kg for the HoCrO₃, Ho_0.67Gd_0.33CrO₃, and GdCrO₃ samples, respectively. Therefore, Gd substitution at the Ho-site was found to considerably improve the magnetocaloric properties of HoCrO₃. Further, present GdCrO₃ bulk sample exhibits giant magnetocaloric properties, which is higher than the reported values for RCrO₃ and RMnO₃ bulk (polycrystalline) samples, making it promising for applications in low-temperature magnetic refrigeration.

This work was funded by the National Science Foundation under Grant No. DMR-1310149.