Introduction

Good magnetic properties, including high Curie temperature and saturation magnetization, make it possible to use lithium ferrites in various microwave devices, in lithium-ion batteries as a cathode material, and in gas sensors [1,2,3,4]. It should be noted that lithium ferrites are of relatively low cost among the whole variety of ferrite materials.

Structural and electromagnetic properties of ferrites depend on both the chemical composition and the synthesis method [5,6,7,8]. Various methods of ferrite synthesis include sol–gel [9 − 11], combustion [12, 13], co-precipitation [14, 15], etc. However, a solid-state method using oxides and carbonates is most widespread for producing lithium ferrites [16, 17]. Here, mechanical activation of the initial oxides can significantly accelerate and reduce the temperature of ferrite synthesis [18,19,20,21,22].

Properties of Li-ferrite substituted by various metal ions such as Ti, Zn, Ni, Mg, Mn, and Co were studied by many researchers [23,24,25,26,27,28,29,30]. This substitution made it possible to obtain ferrites of complex compositions with a specific combination of electrical and magnetic properties. However, ferrites of new compositions with unique properties are currently of increased interest [31].

Recent studies have focused on the properties of various ferrites with rare earth elements. It is known [32] that ions of rare earth elements have unpaired 4f electrons, which provide magnetic anisotropy due to their orbital shape. Magnetocrystalline anisotropy in ferrites is associated with the 4f-3d bond between transition metal ions and a rare earth element. Thus, doping of spinel ferrites with rare earth ions can improve their electrical and magnetic properties. In particular, such changes in the properties of ferrites occur when they are doped with gadolinium (Gd) [33,34,35,36,37,38,39], dysprosium (Dy) [37, 40,41,42,43], neodymium (Nd) [38, 44,45,46], samarium (Sm) [33, 47,48,49,50,51], lanthanum (La) [45, 52,53,54], terbium (Tb) [37, 40, 55,56,57], cerium (Ce) [33, 37, 58, 59], thulium (Tm) [60], erbium (Er) [37, 45, 47, 61, 62], holmium (Ho) [55, 63], ytterbium (Yb) [37], and praseodymium (Pr) [55].

As shown in [34, 35, 40, 64, 65], the introduction of rare earth elements, including Sm, is also of significant effect on the electromagnetic properties of lithium ferrites. In most studies, synthesis of ferrites with rare earth elements and their microstructural characteristics were investigated using X-ray diffraction analysis. It was suggested that substituted ferrites are formed with a small inclusion of secondary phases, for example, GdFeO3 and SmFeO3, depending on the doped rare earth element. Such phases affect the properties of the synthesized ferrites.

As shown in [40, 61, 63, 66, 67], thermal analysis methods provide more reliable data on the synthesis of ferrites with rare earth elements. Therefore, in this study, synthesis of lithium ferrite Li0.5SmxFe2.5-xO4 from mechanically activated precursors Sm2O3/Fe2O3/Li2CO3 was investigated by X-ray diffraction (XRD), thermogravimetric (TG), and differential scanning calorimetric (DSC) analyses. The microstructure of the synthesized ferrite was examined using a scanning electron microscope (SEM).

Experimental

The procedure for preparing powders for synthesis included the following stages:

  • Drying of precursors powders in a laboratory furnace at 200 °C for 180 min;

  • Weighing of Sm2O3/Fe2O3/Li2CO3 powders (the ratio of the phase concentration is given in Table 1) using an analytical balance Shimadzu AUW-D to obtain proportions in accordance with the reactions:

    $$ Li_{2} CO_{3} + 4.88Fe_{2} O_{3} + 0.12Sm_{2} O_{3} \to 4Li_{0.5} Sm_{0.06} Fe_{2.44} O_{4} + CO_{2} \uparrow \;(Sample\;S1) $$
    (1)
    $$ Li_{2} CO_{3} + 4.6Fe_{2} O_{3} + 0.4Sm_{2} O_{3} \to 4Li_{0.5} Sm_{0.2} Fe_{2.3} O_{4} + CO_{2} \uparrow \;\left( {Sample\;S2} \right) $$
    (2)

  • Additional sieving through a metal sieve with a mesh size of 100 mµ;

  • Mechanical activation of powders mixtures in a planetary mill at 300 rpm for 15 min with steel grinding jars and balls. The mixture-to-ball mass ratio was 1:10;

  • Pressing of the mechanically activated mixture using a hydraulic press at 200 MPa for 3 min.

Table 1 Phase concentration of precursors
Full size table

After the above procedure for preparing powders, compacted samples S1 and S2 were synthesized at 900 °C for 240 min in air.

TG/DSC measurements were performed using a STA 449C Jupiter thermal analyzer with corundum crucibles. The heating rate was 10 °C min−1. The formed phases of magnetic ferrite were additionally monitored by TG analysis in a magnetic field, the technique of which is described in [68, 69].

X-ray diffraction was performed using an ARL X'TRA diffractometer with CuKα radiation. Qualitative and quantitative XRD analysis was performed using the PDF-4 + powder database of the International Center for Diffraction Data (ICDD). PowderCell 2.4 software was used to quantify phases, determine crystallite sizes using the Williamson–Hall method, and specify the lattice parameters. For quantitative XRD analysis, a set of experimental data obtained using a diffractometer was modeled by pseudo-Voigt profile function. The ultimate goal of this modeling is to obtain the best theory–experiment approximation.

The microstructure of the samples was investigated using a Hitachi TM-3000 scanning electron microscope.

Results and discussion

Thermal analysis of precursors and synthesized ferrites

In this study, a thermal analysis of the initial oxides used for ferrite synthesis was performed. TG/DSC analysis for iron oxide powder (not shown in this article) revealed no change in mass and significant calorimetric effects up to 800 °C. Only a small endothermic DSC peak at 677 °C associated with the magnetocaloric effect at the Néel temperature of α-Fe2O3 was revealed.

Figure 1 shows the TG and DSC curves for the Sm2O3 and Li2CO3 precursors obtained in the heating mode in a thermal analyzer. For Sm2O3, a three-step mass reduction occurs in the temperature range of 260–460 °C. However, the TG curve shows that the mass change is insignificant and, according to the data reported in [70], it is due to the loss of physically and chemically adsorbed water.

Fig. 1
figure 1

TG/DSC curves of Sm2O3 and Li2CO3 raw reagents at heating regime

Full size image

Thermal analysis of Li2CO3 lithium carbonate indicates that the mass changed near its melting point, which is known to vary within 720–735 °C. Therefore, the peaks on the DSC curve correspond to melting and decomposition of Li2CO3 with the release of carbon dioxide according to the reaction:

$$ {\text{Li}}_{2} {\text{CO}}_{3} {\text{ }} \to {\text{Li}}_{2} O + {\text{CO}}_{2} $$
(3)

The total mass change on the TG curve equal to 57.06% and the enthalpy on the DSC curve equal to 245 kJ mol−1 for the release of CO2 are close to the calculated values of 60% and 230 kJ mol−1, respectively, in accordance with formula (3).

Thermal analysis of Sm2O3/Fe2O3/Li2CO3 precursors mixed in the ratio required to obtain samples S1 and S2 is illustrated in Figs. 2 and 3.

Fig. 2
figure 2

TG/DTG/DSC curves at heating mode (a, c) and cooling mode in magnetic field (b, d) for S1 sample unmilled (a, b) and milled for 300 rpm (c, d)

Full size image
Fig. 3
figure 3

TG/DTG/DSC curves at heating mode (a, c) and cooling mode in magnetic field (b, d) for S2 sample unmilled (a, b) and milled for 300 rpm (c, d)

Full size image

When unmilled powders are heated in the furnace of the thermal analyzer (Figs. 2a and 3a), the TG curve indicates a decrease in mass in several stages. The first stage in the temperature range of 280–450 °C, if compared with Fig. 1, is associated with the mass changes in Sm2O3. The decreased mass (0.43 and 0.86% for samples S1 and S2, respectively) and the enthalpy of the associated heat effect for these samples correspond to the mass content of samarium oxide in the total mixture.

The change in mass in the temperature range of 470–700 °C is associated with the diffusion interaction between the reagents. The total mass loss in this temperature range is 5.01 and 4.59% for samples S1 and S2, respectively. These values are close to the expected ones of 4.92 and 4.64% for CO2 release calculated from Eqs. (1) and (2), respectively. In [18], thermal analysis was used to study the synthesis reactions of lithium ferrites substituted by zinc and titanium. Mass spectrometric analysis of gasses showed only CO2 release in this temperature range. The DSC peak indicates an endothermic peak with an enthalpy of 107 J g−1 for both samples close to the thermal effect of the interaction of iron oxide and lithium carbonate observed in [19].

It is known [18, 19] that mixing lithium carbonate with metal oxides accelerates the process of its decomposition. In particular, in a mixture with Fe2O3, acceleration of the lithium carbonate decomposition is associated with the predominant diffusion of Fe3+ cations into the Li2CO3 crystal lattice. Thus, Fig. 1 shows that the pure lithium carbonate decomposition begins at 700 °C. However, Figs. 2 and 3 illustrate a lower decomposition temperature for Li2CO3 with the release of CO2 in the Sm2O3/Fe2O3/Li2CO3 mixture, which is at least 250 °C lower than that for pure Li2CO3. In this case, the reagents interact in two stages.

The study of the interaction between the Fe2O3 and Li2CO3 reagents to obtain lithium ferrites showed that the limiting stages of the reaction differ depending on the reaction temperature [19, 71]. At 500 °C and above, the reaction is mainly limited by diffusion interactions. However, in this case, the initial reaction product at low temperatures is lithium ferrite LiFeO2, regardless of the ratio of the initial components. With the initial molar ratio of Li2CO3 and Fe2O3 equal to 1:5 and a further increase in the temperature and duration of the synthesis, the reaction product is lithium ferrospinel Li0.5Fe2.5O4.

In this regard, a two-stage change in mass on the TG curve and a double peak on the DSC curve observed in Figs. 2a and 3a at 470–700 °C are associated with the formation of transition phases of lithium ferrite. This is evidenced by a small peak on the DSC curve at 750–760 °C, which is due to the order–disorder (α → b) phase transition in the formed Li0.5Fe2.5O4 phase [18, 72]. The higher enthalpy of 0.96 J g−1 for sample S1 compared to the value of 0.42 J g−1 for sample S2 characterizes the higher ferrite content in sample S1 synthesized at the time of recording the DSC signal. This transition but with a higher intensity due to additional heating can similarly be observed on the DSC curve obtained at the cooling stage (Figs. 2b and 3b). The obtained enthalpy values of 8.9 J g−1 and 8.2 J g−1 for samples S1 and S2, respectively, are very close to the transition enthalpy for pure lithium ferrite with the DSC peak equal to 12 J g−1 [26].

Figures 2b and 3b demonstrate the results of TG analysis of the samples, when the magnets were attached immediately after the sample heating mode in accordance with the scheme proposed in [68]. The TG curves measured using a magnetic field show an abrupt increase in mass near the Curie temperature (Tc), which can be determined from the DTG curves. The Tc for samples S1 and S2 attains 629.2 and 629.7 °C, respectively, and these values are close to the temperature of the magnetic transition in ordered lithium ferrite with Tc = 632 °C [26, 68]. As reported in [72], a slight decrease in the Curie temperature of the samples can be associated with an increase in the degree of the order violation in the ferrite structure. In this case, this is apparently due to the addition of Sm2O3.

Thus, the obtained results indicate the formation of a significant amount of unsubstituted lithium ferrite Li0.5Fe2.5O4 during solid-phase interaction in the unmilled Sm2O3/Fe2O3/Li2CO3 mixture. Based on the results obtained, it is possible to estimate the content of lithium ferrite formed in the samples during thermal analysis. According to the thermomagnetometric methodology of ferrites presented in [68], the mass jump at the Curie temperature was observed to be 0.41% for pure lithium ferrite. As mentioned above, the enthalpy of α − Li0.5Fe2.5O4 to β − Li0.5Fe2.5O4 transition according to the DSC analysis for pure lithium ferrite is 12 J g−1 [26]. Thus, the 0.33 and 0.37% in mass on TG curves as well as 8.9 and 8.2 J g−1 on DSC curves for samples S1 and S2, respectively, indicated that the ferrite content exceeds 70 mass% at the stage of cooling the mixture in a thermal analyzer.

The results of thermal analysis under non-isothermal heating of powders mechanically activated at 300 rpm are presented in Figs. 2c and 3c for samples S1 and S2, respectively. In such samples, a slight decrease in mass can also be observed at 280–450 °C, which is associated with mass changes in Sm2O3. The higher values of mass change observed in this temperature range for both samples (0.46 and 0.99%), compared with the calculated values for the Sm2O3 content in the total mixture, are associated with the overlap of the temperature intervals of the processes occurring in Sm2O3 and the interaction between Fe2O3 and Li2CO3.

The reaction of interaction between the milled Sm2O3/Fe2O3/Li2CO3 precursors with the release of CO2 starts at a temperature of ca. 450 °C. The final temperature of the reaction is 700 °C. The total mass loss in this temperature range is 4.90% and 4.59% for samples S1 and S2, respectively. According to Eqs. (1) and (2), these values within experimental error are close to the calculated ones for the release of CO2. The behavior of the TG and DSC curves indicates that the synthesis reaction proceeds in two-stage with an enthalpy slightly lower than that of the reaction from non-activated powders. In this case, the main mass loss occurs in the first stage of the reaction up to 600 °C, which is 100 °C lower as compared to non-activated powders.

The enthalpies of α − Li0.5Fe2.5O4 to β − Li0.5Fe2.5O4 phase transition, which are equal to 0.99 J g−1 and 0.8 J g−1 for samples S1 and S2, respectively, are higher than the values shown in Figs. 2a and 3a. This indicates the formation of a higher concentration of lithium ferrite in the milled samples during heating in thermal analyzer. Thus, mechanical activation of the reagent mixture accelerates the synthesis reaction between Sm2O3/Fe2O3/Li2CO3 precursors.

The results of TG analysis in a magnetic field presented in Figs. 2d and 3d reveal a high content of Li0.5Fe2.5O4 lithium ferrite in both samples during solid-phase interaction in the mechanically activated Sm2O3/Fe2O3/Li2CO3 mixtures.

For milled sample S1, higher DSC peak of 9.7 J/g and the mass change of 0.4% indicate an increase in the lithium ferrite concentration during sample cooling in the thermal analyzer. However, the lower DSC peak of 6.6 J/g and the mass change of 0.36% for milled sample S2 are apparently associated with a decrease in the amount of lithium ferrite due to the formation of other phase products such as SmFeO3, as shown below.

Figure 4 presents the results of thermal analysis obtained at the cooling stage for samples synthesized at 900 °C for 240 min from mechanically activated powders. According to the results of DSC analysis that show high values of the enthalpy of the β − Li0.5Fe2.5O4 to α − Li0.5Fe2.5O4 phase transition, the synthesized samples contain mainly lithium ferrite Li0.5Fe2.5O4. The content of this ferrite is higher in samples S1 compared to samples S2. No Curie temperatures connected to Li0.5Sm0.06Fe2.44O4 and Li0.5Sm0.2Fe2.3O4 phases were found.

Fig. 4
figure 4

TG/DTG/DSC curves at cooling mode in magnetic field for Sample S1 (a) and S2 (b) synthesized at 900 °C for 240 min

Full size image

X-ray diffraction analysis of synthesized ferrites

For a more comprehensive understanding of the processes that occur during the synthesis of ferrite, we performed XRD analysis of the samples synthesized at 900 °C for 240 min. Figure 5 shows XRD patterns for samples with different levels of doping with Sm3+ ions. It was found that some of the obtained peaks correspond to the cubic space group (Fd-3 m). This indicates the formation of a spinel ferrite structure, namely, an ordered phase α − Li0.5Fe2.5O4 (PDF No. 04–015-5965). Yet, along with ferrite, a small amount of a crystalline secondary phase identified as SmFeO3 is formed (PDF No. 00–039-1490) [65, 73]. It should be noted that SmFeO3 is a non-magnetic material that exhibits an orthorhombic space group (Pnma). This phase can be formed as a result of the following reaction: 5Fe2O3 + 5Sm2O3 = 10SmFeO3.

Fig. 5
figure 5

X-ray diffraction of samples synthesized at 900 °C for 240 min

Full size image

Thus, the results of XRD analysis indicate the formation of a two-phase product based on Li0.5Fe2.5O4 and SmFeO3 in the process of solid-phase interaction. The quantitative content of phases is given in Table 2. At the same time, an increase in the Sm2O3 content from 4.7 to 14.7 mass% in the Sm2O3/Fe2O3/Li2CO3 mixture leads to an increase in the amount of the secondary phase SmFeO3 from 4.9 to 17.5 mass% in synthesized samples. As shown in Table 2, the changed Sm2O3 content causes changes in the lattice constant.

Table 2 X-ray diffraction data of ferrites synthesized at 900 °C for 240 min
Full size table

Thus, for the synthesized samples, XRD and thermal analyses did not reveal the substituted phases of lithium ferrite, such as Li0.5Sm0.06Fe2.44O4 and Li0.5Sm0.2Fe2.3O4.

SEM analysis of synthesized samples

Figure 6 shows SEM images of the samples synthesized at 900 °C for 240 min. They demonstrate two distinct contrasts that presumably correspond to the ferrite phase (gray tint) and SmFeO3 phase (white tint). Thus, the results obtained using SEM analysis confirm the conclusion drawn earlier on the formation of a two-phase product during the synthesis from the Sm2O3/Fe2O3/Li2CO3 powder mixture.

Fig. 6
figure 6

SEM micrographs of samples synthesized at 900 °C for 240 min

Full size image

Conclusions

In this study, we examined the synthesis of lithium-substituted ferrite from mechanically activated Sm2O3/Fe2O3/Li2CO3 precursors using XRD, TG, DSC, and SEM analyses.

It was shown by thermal analysis that mechanical activation of the Sm2O3/Fe2O3/Li2CO3 mixture in a planetary mill increases the powder reactivity, so that the main reaction of reactants interaction occurs at lower temperatures. It was also found that the heating of the Sm2O3/Fe2O3/Li2CO3 mixture leads to the formation of a two-phase product that consists mainly of lithium ferrite Li0.5Fe2.5O4 and the SmFeO3 phase, the amount of which depends on the Sm2O3 content in the initial mixture. The synthesis of pure lithium ferrite was confirmed by XRD analysis data, the Curie temperature point, and by the presence of an endothermic peak on the DSC curve corresponding to the order–disorder transition in the Li0.5Fe2.5O4 phase. The amount of lithium ferrite formed depends on the degree of powder milling. However, regardless of the mechanical activation of the powders, no substituted lithium ferrites such as Li0.5Sm0.06Fe2.44O4 and Li0.5Sm0.2Fe2.3O4 were formed.

It is known that unsubstituted lithium ferrite with the chemical composition of Li0.5Fe2.5O4 is not used in microwave technology due to high dielectric losses. However, this ferrite has the high values of saturation magnetization and Curie temperature. It can be assumed that the formation of Li0.5Fe2.5O4/SmFeO3 composite structure can primarily affect the electrical properties of ferrites. In this regard, this work needs to be continued in terms of studying the electrical and magnetic properties of synthesized ferrites.