Mixed-valence La_0.80(Ag_1−x Sr _x )_0.20MnO₃ manganites with magnetocaloric effect

Abstract

We report the effect of the simultaneous inclusion of Ag¹⁺ and Sr²⁺ cations on the crystal structure and the magnetocaloric effect (MCE) of La_0.80(Ag_1−x Sr _x )_0.20MnO₃ manganites (x = 0.0, 0.25, 0.50, 0.75, 1.0) synthesized by the solid-state reaction method. X-ray photoelectron spectroscopy was used to verify chemical composition and the amount of Mn³⁺–Mn⁴⁺ cations, while the magnetic performance was evaluated by means of a Physical Property Measuring System. The progressive substitution of Ag¹⁺ by Sr²⁺ causes the increase of the number of Mn⁴⁺ cations, together with increasing ionic radii for the A-site contents up to x = 0.75, both favoring the enhancement of the double-exchange interaction and hence, the MCE. Excellent values of magnetic entropy change (−4.6 J/kg K, μ _oΔH = 5.0 T) were observed for the x = 0.25 manganite, while for the x = 0.50 sample interesting refrigerant capacity (129 J/kg, μ _oΔH = 2.0 T) and a wide interval of temperature at the full-width at half maximum δT _FWHM of the magnetic entropy change curve were recorded (101 K, μ _oΔH = 2.0 T) caused by two successive magnetic transitions. Magnetic performance was explained in terms of the effect of the cation substitution on the double-exchange interaction and the tolerance factor.

Introduction

The magnetocaloric effect (MCE) refers to the isothermal magnetic entropy change (or the temperature change under adiabatic conditions) of a magnetic material under the application of an external magnetic field change μ _o ΔH. This phenomenon is known since 1881, however, it was not up to the 1930s, that it was used for reaching ultra-low temperatures (i.e., ≪1 K) by means of the adiabatic demagnetization of paramagnetic salts like Gd₂(SO₄)₃·8H₂O [1]. The MCE has recently elicited a renewed interest due to the possibility of developing magnetic refrigeration technology applicable at room temperature, with the expectation of achieving higher cooling efficiencies (over 50 % of a Carnot cycle, [2–4]), compared with conventional refrigeration based on gas compression of pollutant compounds like CFC. One of the figures of merit for materials characterization aiming to MCE applications is the variation of the magnetic entropy ΔS _M for a given applied magnetic field change μ _oΔH = μ _o (H _f − H _i) (where μ _o H _i is usually zero). In adiabatic conditions, the decrease (increase) of the magnetic entropy due to the application (removal) of the magnetic field leads to the increase (decrease) of the lattice entropy giving rise to a temperature increment (decrement) of the material. An additional parameter for MCE characterization is the refrigerant capacity (RC), which measures the ability of the magnetic refrigerant to transfer thermal energy between the cold (T _cold) and hot (T _hot) sinks when an ideal thermodynamic cycle is considered. To date, the reference material for MCE applications at room temperature is pure Gd, which possesses the most promising combination of properties for magnetic refrigeration applications at room temperature: ΔS _M of −10.2 J/kg K (μ _oΔH = 5 T), RC of 410 J/kg, and T _C = 294 K [4] alongside an adiabatic temperature change ΔT _ad of 6.0 K for μ _oΔH = 2 T [5]. Unfortunately, the considerable high price and cost of production (USD $4000/kg [4]), makes Gd a disadvantageous choice as magnetic refrigerant. Because of this, a wide variety of materials have been explored for MCE applications, including MnFePAs alloys (with ΔS _M as high as −18 J/kg K μ _oΔH = 5 T and T _C = 308 K [6]), MnFePAsGe alloys (with ΔS _M of up to −40 J/kg K, μ _oΔH = 3 T, T _C = 270 K [7]) Heusler alloys like NiMnGa (with ΔS _M of up to −10.4 J/kg K, μ _oΔH = 2 T, and RC = 47 J/kg [8], NiMnCoSn (with ΔS _M = −30 J/kg K, μ _oΔH = 1 T, T _C = 271 K and RC ~ 44 J/kg [9]), or NiMnIn with structural and magnetic entropy changes between −2.6 and −7.2 J/kg K, μ _oΔH = 3 T and RC between 60 and 95 J/kg [10], rare earth-based bulk metallic glasses, like GdAlCo-type alloys characterized by ΔS _M values between −7.6 and −9.4 J/kg K, μ _oΔH = 5 T, and RC 413–679 J/kg, but low Curie temperatures, typically well below 100 K (which constrains refrigerant applications to such temperature intervals) [11] or HoDyAlCo systems with ΔS _M of up to −11.7 J/kg K, μ _oΔH = 5 T, RC = 365 J/kg and T _C = 17 K [12].

An ample group of materials which has been extensively studied for MCE applications are La-based manganites with perovskite structure, because of their competitive cost of production, chemical stability and easy tailoring of magnetic properties [2, 4]. In fact, a significant number of compositional variations have been explored for LaMnO₃-type manganites, including cations with 2+ or 1+ valence states, such as LaSr, LaCa, LaBa, LaPb, LaNd, LaY, LaLi, and La(Na,K,Ag), among others [4, 13–16]. However, reports concerning the effect of concomitant cations with mixed-valence 2+ and 1+ on La-based manganites are much less numerous [2, 4]. In this work, we present a systematic study on the effect of the progressive replacement of Ag¹⁺ and Sr²⁺ cations on the MCE of La_0.80(Ag_1−x Sr _x )_0.20MnO₃ manganites.

Experimental techniques

Mixed-valence polycrystalline La_0.80(Ag_1−x Sr _x )_0.20MnO₃ manganites (x = 0.0, 0.25, 0.50, 0.75, 1.0) were synthesized by means of the solid-state reaction method. Details on the synthetic procedure are described elsewhere [17]. Chemical composition across the compositional series was verified by means of X-ray photoelectron spectroscopy in a VG Microtech Multilab ESCA2000 coupled with a CLAM4 MCD detector with Al-K_α radiation. The elemental chemical analysis was carried out by means of deconvolution of the core-level signals of the XPS spectra. This deconvolution process involves the removal of the background by Shirley method [18, 19]. Then, we calibrate the binding energy (BE) with reference materials (La₂O₃, SrCO₃, Mn₃O₄, and AgO) for which we propose Gaussian–Lorentzian curves with spectroscopic criteria and the sensitivity relative factors reported by Scofield [20]. The fitting of the XPS spectra was performed with the program SDP v4.1 [21]. Magnetization measurements were performed by means of a Physical Property Measuring System equipped with the vibrating sample magnetometer module (PPMS-9T Evercool I, Quantum Design platform). The isothermal magnetic entropy change as a function of temperature ΔS _M(T) was calculated from a set of magnetization isotherms measured each 5 K over the temperature interval 200–400 K. The maximum applied magnetic field change μ _oΔH _max was of 5.0 T. The magnetic entropy variation ΔS _M was calculated through the numerical integration of the Maxwell relation [2, 4]:

$$ \Updelta S_{\text{M}} \left( T \right) = \int_{o}^{{B_{\hbox{max} } }} {\left( {\frac{\partial M}{\partial T}} \right)_{B} dB}. $$

(1)

Results

Details on phase distribution analysis by powder X-ray technique are given in [17], for which the solid solution La_0.80(Ag_1−x Sr _x )_0.20MnO₃ across the compositional series was identified together with minor secondary phases Mn₃O₄ and metallic Ag for the sample with x = 0.0. Decreasing peak intensities of the secondary phases was observed as the Sr concentration increases, with no detectable presence of Ag or Mn₃O₄ for the La_0.80Sr_0.20MnO₃ sample. Very similar phase distribution has been previously described in equivalent La_1−x Ag _x MnO₃ manganites [22–24].

A central issue for the correlation between chemical composition variation and magnetic properties of La-based manganites is the relative amount of Mn⁴⁺ and Mn³⁺ ions present in each sample, as will be described within the “Discussion” section. Firstly, we confirm the Mn divalent state in our samples by the splitting of the Mn 3s core-level signal of the XPS spectra corresponding to the La_0.80(Ag_0.25Sr_0.75)_0.20MnO₃ manganite as example (Fig. 1a). The linear relation between the Mn valence and the exchange splitting energy has been described in [25, 26]. As a second step, we quantify the content of Mn⁴⁺–Mn³⁺ cations by using the XPS high-resolution spectra of the core-level energies for the Mn 2p_1/2 and Mn 2p_3/2, for which Fig. 1b displays such core levels for the La_0.80(Ag_0.50Sr_0.50)_0.20MnO₃ manganite as example, together with the same Mn 2p transition for all the compositions (inset). These core-level signals were used to calculate the Mn ⁴⁺–Mn ³⁺ content by means of a deconvolution process using the peaks located at the following BE: (a) For the Mn 2p_3/2 core-level, peaks at BE = 639.43 eV and its satellite at 642.84 eV, both corresponding to Mn³⁺, while peaks at 640.56 eV and its satellite at 646.13 eV correspond to Mn⁴⁺, (b) For the Mn 2p_1/2 core-level, peaks at 651.05 eV and its satellite at 655.00 eV, both corresponding to Mn³⁺, while peaks at 652.25 eV and its satellite at 657.88 eV correspond to Mn⁴⁺. This assignment of peaks and satellites for Mn³⁺ and Mn⁴⁺ cations based on the same Mn 2p_1/2 and Mn 2p_3/2 core levels has been proposed in [27] for equivalent kind of perovskites, studied for catalysis applications. The Mn⁴⁺ content shown in Table 1 exhibits an increasing tendency with increasing Sr content x as a result of the oxidation of Mn³⁺ atoms to compensate the charge of the incorporated Sr²⁺ ions. These values are consistent with those obtained by other experimental techniques such as redox titrations reported for equivalent La_0.80Sr_0.20MnO₃ manganites (with Mn⁴⁺ content of 34.0 at% for the La_0.80Sr_0.20MnO₃ manganite [28]) as well as for equivalent La_0.7Sr_0.3−x Ag _x MnO₃ (x values upto 0.2) with Mn⁴⁺ content between 29.5 and 34.5 at% for similar Ag concentrations [24]. Complementary, the Mn concentration of the Mn₃O₄ phase was evaluated by means of the Mn 2p_3/2 and Mn 2p_1/2 core levels and the peaks located at BE = 641.40 eV and 653.10 eV, respectively. For this phase, the following concentrations (% elemental) were determined: 6.8, 5.2, 4.6, 4.4, 2.4 for the samples x = 0.0, 0.25, 0.50, 0.75, and 1.0, respectively. This decreasing amount of Mn₃O₄ with increasing Sr content confirms the tendency observed by XRD results [17], except of the x = 1.0 sample, for which XPS indicates that Mn₃O₄ is still forming at a small fraction of 2.4 %. Such fraction is below the XRD detection limit, and thus it was not recorded in the corresponding X-ray diffractogram. For the silver concentration we found two oxidation states for Ag atoms, one associated to Ag⁰ (metallic) with BE = 368.20 eV corresponding to the Ag 3d_5/2 core-level, and the other to Ag¹⁺ for the BE = 367.03 eV for the same core-level. The presence of Ag¹⁺ confirms the incorporation of Ag atoms within the crystal unit cell of the La(AgSr)MnO₃ manganites, while the vanishing content of metallic silver (Table 1) is also consistent with XRD results [17]. No silver atoms were detected for x = 1.0. Details for the Ag composition procedure from the deconvolution of the Ag 3d core-level signal are described in [17, 27, 29].

Fig. 1

a Experimental XPS spectra (circles) of the Mn 3s core-level for the La_0.80(Ag_0.25Sr_0.75)_0.20MnO₃ manganite. The fitting of the signal (dashed line in red) was carried out with peaks (continuous lines) associated with Mn³⁺ and Mn⁴⁺ cations. b High resolution XPS spectra of the Mn 2p core-level for the La_0.80(Ag_0.50Sr_0.50)_0.20MnO₃ manganite. The fitting of the signal is shown as dashed line in red. The inset shows the Mn 2p spectra for the whole compositional series (Color figure online)

Table 1 XPS elemental composition (at%) of Mn³⁺ and Mn⁴ atoms for the manganite series La_0.80(Ag_1−x Sr _x )_0.20MnO₃, according to the deconvolution analysis of Fig. 1

Concerning magnetic properties, thermomagnetization M(T) curves under field cooled and field heating regimes for an applied low magnetic field of 5 mT are displayed in Fig. 2a for the La_0.80(Ag_1−x Sr _x )_0.20MnO₃ manganite series. All the curves display a decreasing behavior with increasing temperature as the magnetic order–disorder transition approaches. A small thermal hysteresis (of up to ΔT = 3 K) is manifested for the manganites with x > 0.0. For the x = 0.50 sample, a step is observed on its corresponding M(T) plot between 250 and 260 K, which can be attributed to a second phase present in a volume fraction below the detection limit of the XRD technique (~4 %). This phase probably corresponds to SrMnO₃, with T _C = 260 K [30], which can remain as an intermediate product of the reaction during the formation of La_1−x Sr _x MnO₃ manganites [31]. A subtle, much less defined step in the M(T) curve within the interval 364–374 K is also visible for the x = 0.75 and 1.0 samples [see Fig. 2a], likely corresponding to the marginal secondary phase La_2/3Sr_1/3MnO₃ (T _C  = 368 K [2, 4]), which could be segregated during the formation of La_1−x Sr _x MnO₃ manganites [31]. According to the thermomagnetization data of Fig. 2a, this phase should have a volume fraction around 4.0 %, which is at the threshold detection of the XRD technique. The Curie temperatures for all the La_0.80(Ag_1−x Sr _x )_0.20MnO₃ compounds were taken at the inflexion point of each curve [Fig. 2b]. T _C determined from Fig. 2b showed an increasing tendency with increasing Sr content x up to x = 0.75 (from 225 K at x = 0.0 upto 336 K at x = 0.75), followed by a noticeable reduction at x = 1.0 (T _C = 324 K). The well defined peak at the magnetic transition observed for all manganites with x > 0.0 reflects a good chemical homogeneity, in contrast with the La_0.80Ag_0.20MnO₃ sample. The magnetic transition temperature of the secondary Mn₃O₄ phase is not visible in Fig. 2 because such temperature is far below 200 K (~43 K [32]). The tendency observed for T _c is in agreement with previous reports concerning equivalent La_0.7Sr_0.3−x Ag _x MnO₃ manganites [24].

Fig. 2

a Thermomagnetization M(T) curves for the La_0.80(Ag_1−x Sr _x )_0.20MnO₃ manganite series under field cooled (FC, open symbols) and field heating (FH, full symbols) regimes. The applied magnetic field was fixed at 5 mT for all the samples, b dM/dT curves as a function of temperature for T _C determination

Isothermal magnetization curves up to μ _o H _max = 5 T at variable temperature for selected compositions of the manganites series La_0.80(Ag_1−x Sr _x )_0.20MnO₃ (x = 0.0,0.50,1.0) are displayed in Fig. 3. The transition from ferromagnetic to paramagnetic ordering (i.e., nearly linear-like M(μ _o H) is visible for decreasing temperature for all the samples. No marked deviations were observed for the low-field (<1.0 T) M(μ _o H) curves across the compositional series, which is indicative of the absence of metamagnetic-like transitions [33]. Very similar behavior of the isothermal magnetization was reported for equivalent LaAgMnO₃ manganites prepared by the same method [23, 24]. Arrott plots M ² versus μ _o H/M for selected compositions are shown in Fig. 4 (x = 0.25,0.75). The positive slope observed for these plots is indicative of a ferromagnetic–paramagnetic transition of second-order character [34] and is in agreement with previous reports on equivalent La_0.7Sr_0.3−x Ag _x MnO₃ manganites [24].

Fig. 3

Isothermal magnetization curves up to μ_o H _max = 5 T for the manganite samples a La_0.80Ag_0.20MnO₃, b La_0.80(Ag_0.50Sr_0.50)_0.20MnO₃, and c La_0.80Sr_0.20MnO₃

Fig. 4

Arrott plots for the manganite samples a La_0.80(Ag_0.75Sr_0.25)_0.20MnO₃ and b La_0.80(Ag_0.25Sr_0.75)_0.20MnO₃

The magnetic entropy variations |ΔS _M| as a function of temperature, calculated at variable μ_oΔH _max from 1 to 5 T, are displayed in Fig. 5a for the La_0.80(Ag_0.75Sr_0.25)_0.20MnO₃ manganite. These magnetic entropy plots show single-peak behavior across the temperature interval, with |ΔS ^peak_M | values having an increasing tendency with increasing μ _o ΔH. For μ _o ΔH = 5.0 T, an interesting maximum of 4.6 J/kgK is observed, which is comparable with the |ΔS ^peak_M | = 5.6 J/kg K of pure Gd at μ _o ΔH = 2.0 T [35]. Additionally, an interval of temperature at the full-width at half maximum of the ΔS _M(T) curve (denoted as δT _FWHM = T _hot −T _cold) of 35 K is observed for μ _o ΔH = 2.0 T with |ΔS ^peak_M | = 2.4 J/kg K, which is important for establishing wide thermodynamic cycles for the magnetic cooling process. Complementary, the RC has been calculated by means of the three established criteria used in literature: (a) |ΔS ^peak_M | x δT _FWHM (denoted hereafter as RC-1). Here, δT _FWHM coincides with the temperature interval of the thermodynamic cycle; (b) calculating the integral, under the ΔS _M(T) curves between T _hot and T _cold (hereafter denoted as RC-2, i.e., $RC - 2 = \int_{{T_{\text{hot}} }}^{{T_{\text{cold}} }} {[\Updelta S_{\text{M}} (T)]_{\Updelta B} dT}$), and (c) maximum |ΔS _M| x δT ^RC−3 product under the ΔS _M(T) curve (labeled as RC-3; in this case, the temperature interval define different hot and cold temperatures which are been labeled as T ^RC−3_hot and T ^RC−3_cold ) [36]. The field dependence of the RC values for the La_0.80(Ag_0.75Sr_0.25)_0.20MnO₃ manganite are shown in Fig. 5b. A maximum RC-1 of 215 J/kg was observed at μ _oΔH = 5.0 T. For the manganites corresponding to x = 0.75 and 1.0, equivalent behavior was observed for both |ΔS _M| and refrigerant capacities, with maximum values of 3.4 J/kg K and 275 J/kg and 3.1 J/kg K and 304 J/kg, respectively. Results of the magnetocaloric parameters of the series studied are summarized in Table 2. Similar magnetic entropy variation has been reported in equivalent La_1−x Ag _x MnO₃ manganites by means of isothermal magnetization curves [22, 23] and by heat capacity measurements [37] and for equivalent La_0.7Sr_0.3−x Ag _x MnO₃ (x values upto 0.2) [24].

Fig. 5

a Temperature dependence of the magnetic entropy change |ΔS _M(T)| and b magnetic field dependence of the refrigerant capacity values for the La_0.80(Ag_0.75Sr_0.25)_0.20MnO₃ manganite

Table 2 Peak magnetic entropy change |ΔS ^peak_M |, RC-1, RC-2, δT _FWHM , T _hot, T _cold, RC-3, δT ^RC−3, and T _hot ^RC−3 and T _cold ^RC−3 related to RC-3 for the La_0.80(Ag_1−x Sr _x )_0.20MnO₃ manganites series (0.00 ≤ x ≤ 1.00)

In addition, the magnetic entropy variation calculated up to μ _oΔH = 5.0 T for the La_0.80(Ag_0.50Sr_0.50)_0.20MnO₃ manganite is shown in Fig. 6, for which a double-peak behavior is manifested due to the contribution of the SrMnO₃ secondary phase. For this sample, very larger RC-2 values of 239 and 94 J/kg were recorded for μ _oΔH = 5.0 and 2.0 T, respectively (Table 2), together with a large δT _FWHM as high as 101 K at μ _oΔH = 2.0 T and 110 K at μ _oΔH = 5.0 T, respectively. This wide δT _FWHM has the additional advantage of embracing temperature intervals around 300 K (265 ≤ T ≤ 312 K), which renders this specific composition as potential candidate for room temperature magnetic refrigeration devices. A similar behavior of the |ΔS _M| curve was observed for the manganite corresponding to x = 0.0, with a remarkable δT _FWHM of 113 K at μ _oΔH = 5.0 T. Unlike the x = 0.50 sample, both T _hot and T _cold temperatures for the x = 0.0 manganite are well below 300 K. Results are summarized in Table 2.

Fig. 6

Temperature dependence of the magnetic entropy change |ΔS _M(T)| for the La_0.80(Ag_0.50Sr_0.50)_0.20MnO₃ manganite

Discussion

From a fundamental viewpoint, the electronic properties of mixed-valence perovskites relies to a significant extent on the double-exchange interaction, which is afforded by the conversion of Mn³⁺ cations into Mn⁴⁺ ions upon doping the original LaMnO₃ compound with divalent or monovalent cations [38]. On the basis of the electronic configuration of Mn³⁺ (t ³_2g e ¹_g ) and Mn⁴⁺ (t ³_2g e ⁰_g ), it can be elucidated that the presence of Mn⁴⁺ ions allow the hopping of e_g electrons of Mn³⁺ to neighboring Mn⁴⁺ through double-exchange interaction, which in turn mediates ferromagnetic ordering and electronic conduction [38]. Hence, the increasing tendency of the Curie temperature observed for manganites with Sr contents upto x = 0.75 is consistent with the increasing Mn³⁺ to Mn⁴⁺ relative proportions which is reported in Table 1.

A complementary aspect that influences the electronic properties of these kinds of perovskites is the lattice effects caused by doping with 2+ or 1+ cations of different ionic radius, in particular, the variation of the Mn–O–Mn bond angle [38–40]. In this sense, an important criterion for analyzing the crystal structure of perovskite ABO₃ compounds (where A corresponds to a trivalent or divalent cation and B corresponds to Mn for the case of manganites) is the tolerance factor t, which represents the capacity of the perovskite structure to accept A and B ions that allows dodecahedral and octahedral coordination, respectively [38, 40]. Stable crystal structures have representative t values as follows: for the ideal cubic perovskite t = 1.0, while for the interval 0.9 < t < 1.0, the unit cell presents a rhombohedral distortion. For t < 0.9 the structure becomes orthorhombic [40]. In order to quantify t, the following Equation has been established [38, 40]:

$$ \left( {r_{\text{A}} + r_{\text{O}} } \right) = t \cdot \sqrt 2 \left( {r_{\text{Mn}} + r_{\text{O}} } \right) $$

(2)

where r _A, r _O, r _Mn stand for the ionic radii of the cations A, O, and Mn, respectively.

For the present case, and bearing in mind the size and oxidation states of the cations involved (r _La3+ = 1.50 Å, r _Ag+ = 1.42 Å r _Sr2+ = 1.58 Å, r _Mn3+ = 0.72 Å, r _Mn4+= 0.67 Å, and r _O2− = 1.26 Å [41]), we propose the following expressions for the mean radii 〈r _A〉 and 〈r _B〉 for our La_0.80(Ag_1−x Sr _x )_0.20MnO₃ phases:

$$ \langle r_{\text{A}} \rangle = r_{{{\text{La}}^{{ 3 { + }}} }} (0.80) + r_{{{\text{Ag}}^{ + } }} (1 - x)(0.20) + r_{{{\text{Sr}}^{{ 2 { + }}} }} (x)(0.20) $$

(3)

$$ \langle r_{\text{B}} \rangle =\, r_{{{\text{Mn}}^{{ 4 { + }}} }} (1 - x)(0.20) + r{}_{{{\text{Mn}}^{{ 4 { + }}} }}(x)(0.40) + r_{{{\text{Mn}}^{{ 3 { + }}} }} (1 - \left[ {(1 - x)(0.20) + (x)(0.40)} \right]). $$

(4)

By inserting the experimental Mn³⁺, Mn⁴⁺, and Ag¹⁺ atomic contents of Table 1 into Eqs. 2, 3, and 4, the resulting plots are shown in Fig. 7a. The r _A increases up to x = 0.75 as a consequence of the progressive incorporation of large Sr cations into the crystal structure containing Ag atoms. For x = 1.0, r _A decreases due to the absence of Ag atoms. The same r _A tendency was described by Rietveld analysis in equivalent La_0.7Sr_0.3−x Ag _x MnO₃ manganites [24]. In contrast, r _B remains constant across the compositional series. In addition, the tolerance factor t (Fig. 7b) exhibits an increasing behavior up to the Sr content x = 0.75, reflecting the tendency to a more cubic-like structure with Mn–O–Mn angles (Θ) becoming closer to 180°. The augmenting Θ angles favors DE interaction, which leads to the progressive improvement of T _C observed in Fig. 2 [38, 39]. The reduced t observed at x = 1.0 implies a more distorted crystal structure compared with x = 0.75, which affects the DE yielding reduced Curie transitions. This correlation between structural variation and magnetic properties has been described with profusion in similar La-based manganites [38, 39].

Fig. 7

Ionic radius for the A-site, r _A and the B-site, r _B (a), and tolerance factor of the perovskite structure ABO₃ (b) as a function of the Sr content in La_0.80(Ag_1−x Sr _x )_0.20MnO₃ manganites series

The enhancement of the DE interaction also benefits the MCE performance of the La_0.80(Ag_1−x Sr _x )_0.20MnO₃ manganites series, since improved DE favors higher magnetizations and steeper M(T) curves at the Curie transition, both of which have a beneficial effect on the magnetocaloric parameter |ΔS ^peak_M |. According to Eq. (1), |ΔS _M| is directly influenced by dM/dT. Thus, following the derivatives shown in Fig. 2b, a maximum in |ΔS _M| should manifest for the x = 0.25 sample having the highest slope with well defined shape at the Curie transition, together with a marked reduction for x = 1.00 with a significantly reduced steep at T _C and intermediate values for the remaining compositions. This tendency is confirmed in Fig. 8 for peak magnetic entropy values |ΔS ^peak_M | as a function of the Sr content x for μ_oΔH = 5 T, for which the noticeable enhancement of |ΔS ^peak_M | = 4.6 J/kg K observed at x = 0.25 is consistent with the pronounced steep of its corresponding M(T) curve at the Curie point.

Fig. 8

Peak magnetic entropy values |ΔS ^peak_M | as a function of the Sr content for the La_0.80(Ag_1−x Sr _x )_0.20MnO₃ manganites series (μ_oΔH = 5.0 T)

Conclusion

Mixed-valence La_0.80(Ag_1−x Sr _x )_0.20MnO₃ manganites with concomitant Ag¹⁺ and Sr²⁺ ions were successfully synthesized by the solid-state reaction method. The mixed-valence and content of Mn cations were verified by means of XPS technique, which indicated an increasing number of Mn⁴⁺ ions with increasing Sr content up to x = 0.75. The MCE parameters (|ΔS ^peak_M |, RC, δT _FWHM) showed a clear dependence with the Sr content as a consequence of the progressive improvement of the double-exchange interaction promoted by the increasing Mn⁴⁺ cations and the tolerance factor. Maximum |ΔS _M| of 4.6 J/kgK was obtained for the La_0.80(Ag_0.75Sr_0.25)_0.20MnO₃ manganite at μ_oΔH = 5.0 T.