Structural and magnetic study of Sm doped NiFe₂O₄ nanoparticles

Abstract

NiFe_2−xSm_xO₄ (x = 0.0, 0.01, 0.025, 0.05, 0.075, 0.1) nanoferrites samples were prepared by sol–gel method. The structural properties of the samples were studied by X-ray powder diffraction, where the formation of a cubic spinel phase was confirmed. The lattice parameter and the particle size were determined by applying Rietveld refinement. A slight change in the lattice parameter was observed for Sm doped nanosamples. Whereas, the particle sizes were found to decrease with increasing Sm concentration (x) and have the highest value (42 nm) for pure NiFe₂O₄ nanoparticles. The magnetic properties were characterized using Mössbauer effect (ME) spectroscopy and vibrating sample magnetometer (VSM). ME spectroscopy was used to study the distribution of cations in tetrahedral (A) and octahedral [B] sites of the nanoparticles and the degrees of inversion values were obtained. The saturation magnetization, coercivity and remanent magnetization were determined using VSM measurements. These magnetic parameters vary with Sm substitution and found to be mainly affected by the change in the cations distribution and particle sizes.

1 Introduction

Rare earth (R) doped nano-ferrites have been investigated because of their novel physics and chemical properties. These (R) doped nano-ferrites are technologically important materials that are used to produce permanent magnets, electronic and microwave devices, magnetic fluids, spintronics, data storage, magnetic filters, transformers and magnetic switches [1,2,3,4]. The magnetic properties of these spinel ferrites are governed by the chemical composition and the method of preparation [5,6,7]. In the case of nickel ferrites, altering the chemical composition can be achieved by doping with different transition metals and rare earth (R) ions [5, 8].

Not only is the type of the dopant but also its amount is critical in changing the cation distribution and accordingly the physical properties of the doped Ni-ferrite. Using different synthesizing methods could yield different particle sizes, particle size distribution, morphology and cation distribution [8, 9]. There are many preparation methods of nickel ferrite nanocrystals; mechanical milling, sol–gel methods, ceramic method, microwave processing approaches and co-precipitation, microemulsions. Among these methods, sol–gel process stands as an important method used to obtain ferrite nanoparticles with enhanced homogeneity and controlled particle size in a relatively short processing time and low reaction temperatures.

Single phase NiFe_1.99R_0.01O₄ (R = Sm, Gd, Eu, La) samples were prepared using the standard ceramic method [8]. Mössbauer effect (ME) measurements suggested that R ions substitute Fe³⁺ ions in the octahedral B site. An increase in the coercivity and saturation magnetization was observed for all doped samples, where the highest saturation magnetization was obtained for Sm-doped Ni-ferrite sample. Then, Zn and Ga doped Ni_1−xZn_xFe_1.49Ga_0.5Sm_0.01O₄ samples were prepared using the standard ceramic method [10]. Measurements confirmed that Sm³⁺ ions favor the B sites and support magnetism in the studied system, whereas Zn substitution clearly destabilizes the magnetic order and suppresses magnetism in the studied systems. Later, ultrasmall Zn-substituted Ni–Sm–Ga (NiFe_1.99Sm_0.01O₄ and Ni_1−xZn_xFe_1.49Ga_0.5Sm_0.01O₄) ferrites nanoparticles were synthesized using sol–gel method [9]. The particle size ranged from D = 4–8 nm, and hence a pronounced superparamagnetism was observed for all the prepared nanoparticles.

In this work, Sm-substituted Ni ferrites NiFe_2−xSm_xO₄ (x = 0.0, 0.01, 0.025, 0.05, 0.075, 0.1) were prepared by the sol–gel technique. The influence of Sm³⁺ doping on the structural and magnetic properties of Ni ferrite were systematically studied. The crystal structure was investigated using X-ray powder diffraction and Fourier transform-infrared (FTIR), whereas the cation distribution and the magnetic properties were studied using ME spectrometer and vibrating sample magnetometer (VSM).

2 Experimental

NiFe_2−xSm_xO₄ (x = 0.0, 0.01, 0.025, 0.05, 0.075, 0.1) nanoparticles were synthesized using Fe(NO₃)₃·9 H₂O [99% Merck, Germany], Ni(NO₃)₂·6H₂O [99.9% Strem Chemicals, USA], Sm(NO₃)₃·6H₂O [99.9% Strem Chemicals, USA], citric acid (C₆H₈O₇·H₂O) [99.7% Winlab, UK] and distilled water. The structure of the prepared samples was studied at room temperature using X-ray diffraction (XRD) (Philips X’pert multipurpose diffractometer) with CuK_α radiation (with λ = 1.5418 Å). The crystal structure was refined applying Rietveld profile method using the Rex software [11]. The crystallographic parameters: lattice constant (a), ionic radius per molecule of the tetrahedral and octahedral sites (r_A, r_B), bond length in the two crystallographic sites (R_A and R_B) and oxygen position parameter (u) have been calculated. Jasco FT-IR 310 spectrophotometer were used to carry out FTIR measurements in the frequency range of 2000–250 cm⁻¹. Austin Science Mössbauer Effect Spectrometer with constant acceleration laser-interferometer-controlled drive and data acquisition system is used in a standard transmission setup with a personal Computer Analyzer (PCA II-card with 1024 channel). The radioactive source is ⁵⁷Co embedded in Rh matrix with initial activity of 50 mCi. Metallic iron spectrum is used for the calibration of both observed velocities and hyperfine magnetic fields. The cation distributions and all (ME) parameters are calculated from the recorded Mössbauer spectra. Magnetic measurements on the synthesized nanoferrites were carried out using VSM (9600-1 LDJ, USA) with a maximum applied field of nearly 20 kG at room temperature and the saturation magnetization M_S and the coercivity H_C were determined.

3 Results and discussion

3.1 XRD measurements

The XRD patterns of NiFe_2−xSm_xO₄ (x = 0.0, 0.01, 0.025, 0.05, 0.075, 0.1) nanoparticles are shown in Fig. 1. For all samples, the observed diffraction peaks can be indexed to NiFe₂O₄ with Fd-3m space group. The diffraction patterns exhibit additional impurity peaks of Ni. Using Rex software (Fig. 2), the percentage of Ni phase was calculated and ranged from 3.4% for pure Ni ferrite to < 1% for all Sm doped samples. The lattice parameter a_exp obtained from the Rietveld refinement of XRD measurements is summarized in Fig. 3b. On the other hand, the lattice parameter a_th can be theoretically calculated using the cation distribution obtained from ME spectroscopy (Sect. 3.2). The Mean ionic radius per molecule of the A-site (r_A) and B-site (r_B) are given by [12]:

$${r_A}\;=\;{{\text{C}}_{{\text{Fe}}}}r\left( {{\text{F}}{{\text{e}}^{3+}}} \right)\;+\;{{\text{C}}_{{\text{Ni}}}}r\left( {{\text{N}}{{\text{i}}^{2+}}} \right)$$

(1)

$${r_B}\;=\;\frac{1}{2}\left[ {{{\text{C}}_{{\text{Fe}}}}r\left( {{\text{F}}{{\text{e}}^{3+}}} \right)\;+\;{{\text{C}}_{{\text{Ni}}}}r\left( {{\text{N}}{{\text{i}}^{2+}}} \right)\;+\;{{\text{C}}_{{\text{Sm}}}}r\left( {{\text{S}}{{\text{m}}^{3+}}} \right)} \right]$$

(2)

where r(Fe³⁺), r(Ni²⁺) and r(Sm³⁺) are the ionic radii of Fe³⁺, Ni²⁺ and Sm³⁺ ions given in Table 1. The concentration of Fe³⁺, Ni²⁺ and Sm³⁺ ions in A- and B- sites are C_Fe, C_Ni and C_Sm, respectively and were taken from (ME) results. The average (A–O) and (B–O) bond lengths are R_A and R_B respectively and can be expressed as:

$${R_A}\;=\;{r_A}\;+\;{r_O}$$

(3)

$${R_B}\;=\;{r_B}\;+\;{r_O}$$

(4)

where r_O= 1.32 Å is the radius of oxygen ion. The lattice parameter a_th can be obtained from:

$${a_{{\text{th}}}}\;=\;\frac{8}{3}\left[ {\left( {{r_A}\;+\;{r_O}} \right)\;+\;\sqrt 3 \left( {{r_B}+{r_O}} \right)} \right]$$

(5)

Fig. 1

XRD patterns of different NiFe_2−xSm_xO₄ (x = 0.0, 0.01, 0.025, 0.05, 0.075, 0.1) nanoparticles

Fig. 2

Rietveld refinement profile for NiFe_1.95Sm_0.05O₄ sample

Fig. 3

a Variation of particle size with the substitution of Sm content. b Variation of the lattice parameter (a) with Sm content x. a_exp is the lattice parameter obtained from the XRD measurements and a_th is the lattice parameter calculated according to the cation distribution obtained from the ME measurements

Table 1 The effective ionic radii of r(Fe³⁺), r(Ni²⁺) and r(Sm³⁺) ions in the tetrahedral A-site and octahedral B-site [13]

The calculated values of a_th, r_A, r_B, R_A, R_B and u are illustrated in Table 2. The comparison between the calculated lattice parameter a_th and the experimentally obtained a_exp are shown in Fig. 3b. According to the ME measurements (Sect. 3.3), nano NiFe₂O₄ is a partial inverse spinel, where most of the Ni²⁺ ions set on the B-site, while a small fraction (~ 9%) enters the A-site. By increasing x, Sm³⁺ ions have a clear preference to inter the B-site and gradually substitute Fe³⁺ ions. This process is accompanied by the migration of the Ni²⁺ from the A-site to the B-site. The ionic radius r(Sm³⁺) is bigger than both r(Fe³⁺) and r(Ni²⁺). This why by increasing x, both r_A and R_A slightly decrease, while r_B and R_B increase monotonically. The resultant effect is an increase in the lattice parameter. The values of the lattice parameter are very close to the values observed for the bulk Ni ferrite sample [4]. For the A-site, center of symmetry is at $\left( {\frac{3}{8}\frac{3}{8}\frac{3}{8}} \right)$ and the ideal value of the oxygen positional parameter (u^43m) is 0.375. While, the center of symmetry for the B-site is at $\left( {\frac{1}{4}\frac{1}{4}\frac{1}{4}} \right)$ and the ideal value of u^3m is 0.25. In this case, the arrangement of O²⁻ ions correspond exactly to a cubic closed packing. However, in the actual spinel lattice a small deformation from this ideal pattern is observed. As shown in Table 2, the values of u^3m and u^43m are higher than the ideal values, which indicate that O²⁻ ions move away from the tetrahedrally coordinated A-site cation along the $\langle 111\rangle$ directions due to the contraction of the tetrahedral interstices, correspondingly the octahedral B-sites become smaller [9]. Such changes in the oxygen positional parameter will affect the A–A interactions and B–B magnetic interactions.

Table 2 Experimental (a_exp) and theoretical (a_th) lattice constants, ionic radii of A- and B-sites (r_A and r_B), the average bond lengths (R_A and R_B) and the oxygen positional parameter (u) of NiFe_2−xSm_xO₄ (x = 0.0, 0.01, 0.025, 0.05, 0.075, 0.1) nanoparticles

The particle size of NiFe_2−xSm_xO₄ nanoparticles can be obtained from the Rietveld refinement of XRD measurements. The broad diffraction peaks observed for all the samples reflect the reduced particle size. The particle size D are shown in Fig. 3, where it varies from ~ 42 nm for pure Ni-ferrite to ~ 30 nm for NiFe_1.9Sm_0.1O₄. Generally, by increasing the Sm content x, the particle size decreases. A similar behavior was observed for Nd³⁺ substituted nano Mg–Cd ferrites [14, 15].

3.2 FTIR analysis

FTIR spectra of NiFe_2−xSm_xO₄ (x = 0.0, 0.01, 0.025, 0.05, 0.075, 0.1) nanoferrites are shown in Fig. 4 and the vibration frequencies of all the samples are listed in Table 3. Spinel ferrites generally are branded with two vibrational frequency bands, ʋ₁ around 600–500 cm⁻¹ which is related to the Fe³⁺–O²⁻ and Ni²⁺–O²⁻ complexes at the tetrahedral site (A) and vibrational frequency ʋ₂ around 450–350 cm⁻¹ which is related to the Fe³⁺–O²⁻, Sm³⁺–O²⁻ and Ni²⁺–O²⁻complexes at octahedral (B) site [16]. The differences in vibration frequencies between the two crystallographic A and B-sites are attributed to the differences in bond length between the two sites, where the ionic radius of A-site is smaller than B-site. In this case, the change in values of ʋ₁ and ʋ₂ with increasing Sm content reflects the change in bond lengths of Fe³⁺, Sm³⁺ and Ni²⁺ metal ions complexes present in this ferrite. As discussed before in Sect. 3.1, XRD measurements suggest a slight change in the average bond length R_A, R_B and the lattice parameter a. Accordingly, a minor change in the vibrational frequency ʋ₁ and ʋ₂ was observed. The bands at 1613 cm⁻¹, 1359 cm⁻¹ and 1115 cm⁻¹ are attributed to; (a) the stretching vibration in the residual H–O–H modes of water molecules, (b) the symmetric stretching of the carboxyl group (COOH) and (c) the asymmetric stretching vibration of the adsorbed NO₃ produced as a residue from the nitrate group after the combustion of the samples [17].

Fig. 4

FTIR spectra of different NiFe_2−xSm_xO₄ (x = 0.0, 0.01, 0.025, 0.05, 0.075, 0.1) nanoparticles

Table 3 Tetrahedral and octahedral vibrational frequencies (ʋ₁ and ʋ₂) of NiFe_2−xSm_xO₄ (x = 0.0, 0.01, 0.025, 0.05, 0.075, 0.1) nanoparticles

3.3 ME measurements

Figure 5 shows the room temperature transmission ME spectra of NiFe_2−xSm_xO₄ (x = 0.0, 0.01, 0.025, 0.05, 0.075, 0.1). ME spectra spot the hyperfine interactions between the nucleus and its neighboring electrons. There are three main hyperfine interactions, one originates from the electron density at the nucleus (Isomer shift δ), the other from the gradient of the electric field at the nucleus (Quadrupole splitting ΔE_Q), and finally from the unpaired electron density at the nucleus (Hyperfine field H_f). The spectra of all samples were fitted with two magnetic sextets and a paramagnetic doublet. All the fitting parameters are summarized in Table 4. The magnetic sextets are a direct proof of the magnetic order in the studied samples. The existence of two sextets is typical for inverse and partially inverse spinel ferrites, where they refer to Fe³⁺ in both tetrahedral and octahedral environments. For this reason, the area of these two sextets are used for the determination of Fe amount in the A- and B-sites. The sextets attributed to the Fe³⁺ ions at the octahedral (B) have a larger isomer shift and hyperfine field [18]. As presented in Fig. 6, the Fe³⁺ amount in the B-site is slightly higher than that in the A-site, which straight forwardly refers that a portion Ni²⁺ ions set on A-site (~ 9%). This unusual partial inverse Ni ferrite is due to the reduced particle size [4]. For x = 0.01, Sm³⁺ ions replace Fe³⁺ in the B-site and in the same time some of the Ni²⁺ ions transfer from the A- to the B-site. By increasing the Sm content x, the behavior continues and the amount of Fe³⁺ ions in the A-site exceed that in the B-site for x > 0.05.

Fig. 5

Room temperature ME spectra of different NiFe_2−xSm_xO₄ (x = 0.0, 0.01, 0.025, 0.05, 0.075, 0.1) nanoparticles

Table 4 The isomer shift (δ), quadrupole splitting (ΔE_Q), Hyperfine field (H_f) and the doublet linewidth ω_d of different NiFe_2−xSm_xO₄ (x = 0.0, 0.01, 0.025, 0.05, 0.075, 0.1) nanoparticles

Fig. 6

Fe content calculated from ME spectra for different NiFe_2−xSm_xO₄ (x = 0.0, 0.01, 0.025, 0.05, 0.075, 0.1) nanoparticles

In addition to the two sextets components, the spectra of all samples have doublets due to the presence of nano-sized ferrite crystallites below the single domain size (D_sp) with superparamagnetic behavior [4, 19]. The area of the doublets is ranging from (3.58–16.24%). δ values are typical for Fe³⁺ ions in octahedral and tetrahedral environment [4, 5]. Since, the s-electrons charge density of the Fe³⁺ ions is not influenced by crystalline size, the observed values of the isomer shift show a weak change with decreasing D. The quadrupole splitting (ΔE_Q) has values close to zero for both (A)- and [B]-sites.

Figure 7 presents the hyperfine field (H_fA and H_fB) for both A- and B-sites. Generally, the hyperfine field reflects the internal magnetic field produced by the magnetic ordering on Fe³⁺ ions on both sites. This field is strongly affected by intersite A–A and B–B interactions and the interasites A–B interactions. As the cation distribution changes due to the Sm³⁺ doping, these magnetic interactions change and consequently affect both hyperfine fields. Also, the hyperfine field is sensitive to the decreased particle size, where decreasing the particle size tends to decrease the hyperfine field [20, 21]. Figure 7 suggests that increasing Sm³⁺ concentration slightly changes H_fA and H_fB. Generally, for x ≤ 0.1, Sm³⁺ ions has a minor effect on the magnetic interactions and magnetic order. However, the changes in the cation distribution due to Sm doping is reflected in the difference between H_fB and H_fA. The difference between H_fB and H_fA increases gradually by introducing Sm³⁺ ions and it peaks at x = 0.05 and then decreases again for higher concentration (x > 0.05) where the Fe³⁺ in the A-site exceeds that in the B-site. The change in the cation distribution can also be spotted in the doublets parameters. As indicated in Fig. 8, pronounced differences are observed in the behavior of δ_d, ΔE_Qd and doublet linewidth ω_d for x < 0.05 and x > 0.05. Such behavior can be attributed to the changes in the A–B interactions due to partial transfer of the Fe³⁺ ions from the B-site to the A-site by increasing Sm³⁺ concentration.

Fig. 7

Mössbauer hyperfine field of different NiFe_2−xSm_xO₄ (x = 0.0, 0.01, 0.025, 0.05, 0.075, 0.1) nanoparticles

Fig. 8

For different nanoparticles a Isomer shift of the paramagnetic doublet. b Quadrapole splitting of the paramagnetic doublet. c Linewidth of the paramagnetic doublet

3.4 Magnetic properties

Figure 9 shows the variation of magnetization as a function of the applied magnetic field for various NiFe_2−xSm_xO₄ (x = 0.0, 0.01, 0.025, 0.05, 0.075, 0.1) nanoparticles at 300 K. The M–H curve provides information regarding the magnetic parameters such as saturation magnetization (M_S), coercivity (H_c) and remanent magnetization (Mr), which are summarized in Table 5. Saturation magnetization (M_S) and Coercivity (H_c) are shown in Fig. 10. Generally, the magnetic properties should mainly depend on Sm³⁺ content and the cation distribution of different ions between A and B sites. However, the magnetic properties of nanoferrites also strongly depend on the particle size and the particle size distribution. The saturation magnetization (M_S) is expected to decrease with decreasing the particle size [4]. The magnetic moment per formula unit in Bohr magneton (µ_B) can be calculated using the relation [8]:

$${M_{obs}}\;=\;\frac{{\left( {{\text{saturation~magnetization}}} \right)\; \times \;\left( {{\text{molecular~weight}}} \right)}}{{5585}}$$

(6)

The value of M_obs is presented in Table 4 and Fig. 10a. The magnetic properties of ferrites are mainly governed by super exchange interactions and therefore changing the cation distribution and introducing rare earth Sm would affect A–A, A–B and B–B magnetic interactions. According to Néel’s theory, the magnetic moment M_cal is expressed as:

$${M_{{\text{cal}}}}\;=\;{M_B}\; - \;{M_A}$$

(7)

Fig. 9

VSM measurements for NiFe_2−xSm_xO₄ (x = 0.0, 0.01, 0.025, 0.05, 0.075, 0.1) nanoparticles

Table 5 Saturation magnitization (M_S), Coercivity (H_C), Remanent magnetization (M_r), Squareness ration (R²), observed magnetic moment (M_obs) and calculated magnetic moments according to Néel’s theory (M_cal) for NiFe_2−xSm_xO₄ (x = 0.0, 0.01, 0.025, 0.05, 0.075, 0.1) nanoparticles

Fig. 10

a Comparison between observed magnetic moment obtained from VSM measurements (M_obs) and calculated magnetic moments according to Néel’s theory (M_cal). b Coercivity (H_c) and c Saturation magnetization (M_S) for NiFe_2−xSm_xO₄ (x = 0.0, 0.01, 0.025, 0.05, 0.075, 0.1) nanoparticles

The concentration of different ions in the A- and B- sites are obtained from the ME measurements and the magnetic moment of Ni²⁺, Fe³⁺ and Sm³⁺ ions are 2 µ_B, 5 µ_B and 5 µ_B, respectively. The dashed line is the expected value for complete inverse NiFe₂O₄, which is exact 2 µ_B. For all the samples, M_cal values are significantly higher than the observed values. M_S value depends on the ions magnetic moments, and the distance and angle between two adjacent magnetic ions. The larger ionic radii of Sm³⁺ ions compared to Fe³⁺ ions causes changes in the lattice parameter, bond lengths, bond angles and oxygen positional parameter, which would lead to the variation of the magnetic interactions. Moreover, the change in the cation distribution will also affect the magnetic interactions. Finally, the reduced particle size of the prepared nanoparticles decreases M_S values [3, 22, 23]. The coercivity of the ferrites nanoparticles depends on internal micro stress due to distortion of lattice, different electronic configuration of the ions and particle size. Sm³⁺ ions concentration would logically affect H_C since Sm³⁺ 4f electrons have a strong spin–orbit coupling [22]. The change in coercivity with Sm³⁺ ions concentration can be attributed to the decrease in particle size and the changes in the cation distribution. The squareness R² = M_R/M_S for all the studied samples is ranging from 0.235–0.314. Since the squareness is smaller than 0.5, NiFe_2−xSm_xO₄ nanoparticles would have uniaxial anisotropy which arises due to surface effect [23, 24].

4 Conclusion

NiFe_2−xSm_xO₄ (x = 0.0, 0.01, 0.025, 0.05, 0.075, 0.1) nanocrystalline ferrites were synthesized by sol–gel technique. The structural parameters; lattice parameter (a), bond lengths and oxygen positional parameter (u) were calculated from the X-ray powder diffraction measurements. These structure parameters vary systemically with the substitution of Sm³⁺ ions. ME spectroscopy was utilized to study the magnetic properties and determine the cation distribution. For all the samples, ME spectra contain two magnetic sextets and a paramagnetic doublet, which suggest the coexistence of the magnetic order for large particle sizes and superparamagnetic behavior for small particle sizes below D_sp. The saturation magnitization (M_S), coercivity (H_C) and remanent magnetization (M_r) were determined from the VSM measurements. The squareness ratio suggests that all the studied nanoparticles have uniaxial anisotropy. The observed magnetic moment values (M_obs) for all samples are significantly lower than the calculated magnetic moments according to Néel’s theory (M_cal). These reduced values of M_obs can be attributed to the reduced particle sizes of the prepared nanoferrites.