Revealing the Impact of Ga and Y Doping on Thermal and Electrical Behavior of LaMnO₃ Ceramic Materials

Abstract

The synthesis, thermal behavior and electrical properties of a series of undoped and 1% Ga- or Y-doped lanthanum manganite compounds, obtained via the sol–gel technique, are reported. Scanning electron microscopy (SEM/EDX), X-ray diffraction (XRD) and Fourier transform infrared spectroscopy (FT-IR) analyses were performed on all synthesized samples. The XRD results confirmed a good crystallinity for all studied samples, and a change in the crystal structure of Ga- or Y-doped lanthanum manganite (Pm-3m space group) was observed compared to the pristine sample (R-3c space group). Thermal analysis highlighted a different behavior of the doped samples compared to the undoped sample, observed by the different mass losses in the analyzed temperature range. For these materials, it is shown for the first time that the static electrical conductivity, σ_DC, of Ga- or Y-doped LaMnO₃ compounds increases compared to the σ_DC of the pristine sample, and the thermal activation energy of the process of electrical conduction, E_A,cond, increases linearly with the temperature for all three studied samples.

1. Introduction

For over seven decades, research has focused on manganese oxides exhibiting a perovskite-type structure, characterized by the chemical formula AMnO₃ (where A = Ca, La, Sr or Ba) [1]. Jonker and Van Santen were pioneers in this field in 1950, being the first to synthesize mixed-valence manganite (La_1-xCa_xMnO₃) [2].

Manganese–rare earth hybrid oxides have been investigated regarding the interplay between their orbital, lattice, magnetism and transport properties in various R_1-xA_xMn _1-yByO₃ compounds (where A is a divalent cation, B is a transition metal and R is a rare earth metal). Mn ions in the LaMnO₃ system are trivalent, and super exchange interaction exists through Mn⁺³–O–Mn⁺³ pathways [3]. Lanthanum manganese’s materials present remarkable structural and electromagnetic properties [4,5,6]. Furthermore, doped LaMnO₃ compounds have attracted considerable attention due to their improved properties required for different application domains. These compounds have also gained recognition as appropriate cathode materials for fuel cells due to their excellent electrical conductivity and chemical stability, especially at high temperatures [7].

Many techniques can be applied to synthesize undoped and doped LaMnO₃ materials, such as sol–gel, hydrothermal, solid-state reaction and combustion methods, or the ultrasonic method, which employs a sonotrode immersed in a reaction medium [8,9,10,11,12,13,14].

By doping, the semiconductor’s properties are enhanced by adding heteroatoms into a target lattice without changing the host crystal’s basic characteristics and structure. For example, in the case of Ga-doped LaMnO₃ perovskite materials synthesized by a conventional ceramic procedure, the increase in the Ga content determines a decrease in the transition pressure to the higher-symmetry phases [13]. Previous reports on perovskitic materials doped with various rare earth and transition metals indicate the influence of dopants on a material’s properties. For example, electrical conductivity is influenced by both the ionic radius of the dopant and its concentration, as shown in [14], where, for the studied perovskite materials, the electrical conductivity increased as the ionic radius of the used dopants decreased [14]. In this context, the detailed study carried out in the work herein is based on the structural, thermal and electrical properties of undoped and Ga- or Y-doped lanthanum manganite synthesized by the sol–gel technique. The dopants’ influence on the structural properties of LaMnO₃ perovskite materials is investigated, and for the first time, their effect on the electrical properties is shown through DC electrical conductivity measurements, where Ga or Y ions in the LMO perovskite sample lead to a decrease in the E_A,cond. Moreover, the VRH (variable-range hopping) model of Mott was applied to explain the mechanism of conduction in these samples, and comparisons of the Mott parameters were computed.

2. Materials and Methods

2.1. Materials and Reagents

Ga- and Y-doped LaMnO₃ and pure LaMnO₃ were synthesized using the sol–gel method [5], and then a low temperature heat treatment was performed. The starting chemical reagents were as follows: La(NO₃)₃•6H₂O, Mn(NO₃)₂•4H₂O, Ga(NO₃)₃•× H₂O/Y(NO₃)₃•6H₂O and 2M NaOH solution. All chemicals reagents used were purchased from Sigma-Aldrich, Saint Louis, Missouri, Statele Unite, with the highest purity (p.a.), and were used without any further purification.

The stoichiometric concentrations for these precursors were first computed, and then the materials were dissolved in a mixture of distilled water and ethyl alcohol. Finally, 1% gallium or yttrium dopant was added, and stirring was continued for another 30 min at 60 °C. In the next step, the temperature was increased to 140 °C and maintained at this level until powders were obtained. The powders thus obtained were thermally treated for 6 h at a temperature of 600 °C.

2.2. Characterization of Materials

In order to confirm the crystal structure of the ceramics, the X-ray diffraction (XRD) analysis of powders was performed (with Cu-Kα radiations, λ = 0.15406 nm) by means of a PANalytical diffractometer- Philips, FEI Company PANalitical BV Netherlands, over the range of 2θ from 15° to 80°, at a 2°/min scan rate. Rietveld refinement method, with the High Score X′Pert Plus program, was employed for crystalline structure analysis, using FullProf software Version: 2.2b (2.2.2).

Thermal analysis was carried out in a dynamic air atmosphere (with a flow rate of 100 mL min⁻¹ of synthetic air 5.0 Linde Gas) to assess the potential thermal-oxidative degradation stability of the samples. Measurements were performed using a Diamond TG/DTA PerkinElmer instrument, Switzerland, employing open alumina crucibles, over the temperature range of 25–900 °C, with a heating rate of 10 °C min⁻¹.

A Shimadzu Prestige spectrometer, Japan was used for FT-IR studies, in KBr pellets, covering the spectral range 500–4000 cm⁻¹. Qualitative analysis of the surface was carried out by scanning electron microscopy, using an Inspect S PANalytical Scanning electron microscopy (SEM/EDX), FEI Company PANalitical Netherlands. The components of the complex impedance, Z= Z′ + i Z″ (where $i=\sqrt{-1}$), of the samples were measured at a constant frequency, f = 1 kHz, and different temperatures situated between 28 and 120 °C by means of a LCR-meter GW-INSTEK LCR-6020 type, in conjunction with an oven, in which a capacitor containing the sample was inserted [15]. The experimental setup is similar to ASTM D150-98 [16].

3. Results and Discussion

3.1. XRD Analysis

The crystal structure analysis through XRD powder diffraction was performed for all three compounds, and as observed in Figure 1, the samples present a very good crystallinity, proven by sharp and well-defined peaks. The perovskite-type structure is characteristic for the undoped LaMnO₃ sample indexed in the R-3c space group (corresponding to JCPDS card file no. 01-089-0678). This structure has also been reported for other LaMnO₃ materials obtained by the sol–gel method [17,18]. The 1% Y³⁺- and Ga³⁺-doped LaMnO₃ structures induced changes in the crystal structure, as some of the peaks on the observed diffractograms are not split compared to the pristine compound (marked with symbols in Figure 1). Therefore, the 1% Ga- and Y-doped LaMnO₃ samples are indexed in the Pm-3m space group (JCPDS card file no. 01-075-0440). The refined unit cell parameters extracted for the perovskite samples using the R-3c space group (undoped LaMnO₃ sample) and the Pm-3m space group (Ga-/Y-doped LaMnO₃ samples) are indicated in

Table 1. The refined unit cell parameters for undoped LaMnO₃ (R-3c space group) and Ga-/Y-doped LaMnO₃ (in Pm-3m space group) samples.

Sample	Space Group	Lattice Parameter	Unit Cell Volume	Crystallite Size
		[Å]	[Å3]	[nm]
LaMnO3	R-3c	a = b = 5.4749(5) c = 13.327	345.9341	25.8
LaMnO3:Ga	Pm-3m	a = b = c = 3.8677(7)	57.8584	17.3
LaMnO3:Y	Pm-3m	a = b = c = 3.8687(4)	57.9007	26.8

.

3.2. Thermal Analysis

The thermal analysis was conducted in a synthetic air atmosphere, over the temperature range of 25–900 °C, and revealed that the thermal behavior differs from one sample to the other. The TG/DTG curves are presented in Figure 2. For the undoped sample, there is a 2.13% loss of mass in the range of 25–689 °C and then a 0.2% loss of mass in the range of 690–820 °C. Also in this last interval, an exothermic process with a ΔH = −29.48 J∙g⁻¹ is observed on the HF curve.

In the case of the Ga-doped LaMnO₃ material, a mass loss of 5.11% (Figure 2) is observed in the range of 25–600 °C and a mass loss of 2.8% is observed over 600 °C. This last decomposition process is accompanied by an exothermic process, visible on the Heat Flow curve, with ΔH = −22.59 J∙g⁻¹. The Y-doped LaMnO₃ shows a higher mass loss than in the undoped and Ga-doped samples. In this case, the mass loss is 7.02% in the range of 25–600 °C. Over 600 °C, an exothermic process takes place with ΔH = −27.15 J∙g⁻¹ and with a mass loss of 3.01% of the sample mass (Figure 2).

3.3. FT−IR Spectroscopy

The FT-IR spectra of powder samples of undoped LaMnO₃ and samples doped with Ga³⁺ or Y³⁺ are presented in Figure 3 over the 400–4000 cm⁻¹ wavenumber range. The FT-IR spectra for the LaMnO₃ samples doped with Ga³⁺ or Y³⁺ ions have the same profile as the pristine sample, and it was found that there are no new bonds for the two dopants’ transition elements. Thus, the most significant absorption bands are 600, 850, 980, 1368, 1480 and 3400 cm⁻¹. The absorption bands recorded around 600 cm⁻¹ for all samples are due to vibrational modes for the Mn-O-Mn bending and Mn-O stretching bonds of the octahedral MeO6 group, associated with the change in length of the metal–oxygen bonds typical for perovskite compounds [19].

The two bands situated at 1368 and 1480 cm⁻¹, respectively, are due to the presence of the CO₂ molecule absorbed on the material’s surface. The broad low-intensity peak with a maximum at approximately 3400 cm⁻¹ may be attributed to two kinds of OH- groups, such as O-H bonds assigned to the stretching modes of surface-adsorbed water and an OH-group corresponding to the lattice connections [20].

3.4. Scanning Electron Microscopy

Figure 4 shows the surface morphology of Ga- or Y-doped perovskite materials and was obtained by scanning electron microscopy. In order to highlight the shapes and sizes of the particles, the studied materials were analyzed in the low-vacuum mode. Thus, from the presented images, one may see that in both cases the particles have a spherical shape, with dimensions of hundreds of nm, being strongly agglomerated. Furthermore, from the images presented in Figure 4, it can be seen that the nature of the dopant did not influence the shape of the particles.

Moreover, it has to be mentioned that quantitative analysis by EDX demonstrates that the samples have the expected/determined elemental composition. The stoichiometry of the obtained samples was confirmed by the quantification of the elemental atomic ratio of La:Mn:O, i.e., 1:1:3. Thus, a homogeneous distribution of the corresponding La, Mn, O and Ga/Y elements in the synthesized samples was observed.

3.5. Electrical Properties

With the measured values of $Z^{\prime }$ and $Z^{″}$, the electrical conductivity σ, can be computed for each temperature, T, using the following equation:

$$\sigma =\frac{Z^{\prime }}{{\left|Z^{\ast }\right|}^2}\cdot \frac{d}{A}$$

(1)

Here, A and d stand for the transversal sectional area and the length of the sample, respectively, and $\left|Z^{\ast }\right|=\sqrt{{Z^{\prime }}^2+{Z^{″}}^2}$ represents the modulus of the complex impedance of the sample. It is known that [21] in the low-frequency range, ($f\le 1\mathrm{kHz}$), the value of σ, determined by relation (1), represents the DC electrical conductivity, ${\sigma }_{DC}(T)$, which does not depend on frequency but on temperature. For each sample, the temperature dependence of static conductivity, σ_DC(T), was plotted by using the conductivity values obtained at the frequency of f = 1 kHz at temperatures ranging from 28 °C to 120 °C. This plot is depicted in Figure 5.

Figure 5 shows that the static conductivity, σ_DC(T), rises with temperature across all samples, demonstrating a thermally activated conduction process within the measurement temperature range of 28–120 °C. The electrical conductivities of LMO:Ga and LMO:Y, respectively, are higher than that of the LMO sample, which shows that the presence of rare earth ions of the Ga³⁺ or Y³⁺ types in the LaMnO₃ perovskite sample leads to the electrical conductivity of these samples increasing compared to the pristine sample (LMO). On the other hand, the substituted Ga³⁺ ions in the LaMnO₃ perovskite crystal lattice behave as acceptor dopants replacing the Mn³⁺ ions in the octahedral B-site [22], which leads to the compensation of the charge imbalance following their oxidation. As a result, gallium doping slightly reduces the octahedral distortion and diminishes the Jahn–Teller effect by enhancing electronic correlations [23,24], resulting in a larger increase in the electrical conductivity [25] of the LMO system doped with Ga ions, as can be observed in Figure 5. The yttrium ions replace the lanthanum ions in the tetrahedral A-site of the LMO perovskite crystal lattice [26], and as a result, the yttrium substitution leads to a lowering of the mean ionic size at the La-site, which leads to distortion of the MnO₆ octahedra. The Mn³⁺–O–Mn⁴⁺ bond angle thus changes from 180°, which leads to a decrease in the probability of electron hopping between Mn³⁺ and Mn⁴⁺ ions [27], and will result in a lesser rise in the electrical conductivity of the LMO system doped with Y ions, as was determined experimentally (see Figure 5).

Also, the rise in σ_DC with temperature stems from the growth of the drift mobility of charge carriers in the sample, in agreement with Mott’s model of variable-range hopping (VRH) [28]. This type of behavior with the temperature of the conductivity σ_DC(T) has also been observed for other perovskite materials, such as NaTaO₃ [29] doped or not doped with metallic ions (Ag, Fe, Cu) [11,30,31], which shows that the most suitable electrical conduction mechanism in these materials is the VRH mechanism. According to the VRH model, the σ_DC(T) conductivity is given by:

$${\sigma }_{DC}={\sigma }_0\exp \left[-\frac{B}{{(T)}^{1/4}}\right]$$

(2)

where σ₀ is the pre-exponential factor and

$$B=\frac{4E_{A,cond}}{k{T}^{3/4}}$$

(3)

In Equation (3), k stands for the Boltzmann’s constant, whilst E_A,cond represents the thermal activation energy of electrical conduction [28]. Using Equation (2) and the σ_DC(T) conductivity values from Figure 5, we plotted the experimental dependence, ln σ_DC (T^−1/4), which is shown in Figure 6.

Slope B of the linear dependence, ln σ_DC (T^−1/4), was found via the linear fitting of the experimental dependencies from Figure 6. Knowing slope B, the thermal activation energy of electrical conduction, E_A,cond, was determined using Equation (3). The temperature dependence of E_A,condn (T) is presented in Figure 7.

From Figure 7, one can observe a linear increase in E_A,cond with the increase in temperature from 0.177 eV to 0.218 eV for sample 1 (LMO), from 0.175 eV to 0.213 eV for the LMO:Ga sample and from 0.172 eV to 0.210 eV for the LMO:Y sample. As a result, the presence of Ga³⁺ or Y³⁺ rare earth ions in the LMO perovskite sample leads to a decrease in E_A,cond corresponding to these samples (LMO:Ga and LMO:Y) as compared to the LMO sample. Thus, the increase in conductivity of the doped materials is determined in relation to the electrical conductivity of the undoped sample, as obtained experimentally (see Figure 5). The electrical conduction mechanism of the study samples can be described in terms of the process of charge carriers hopping between the localized states [28,32]. σ_DC(T) may also be expressed as:

$${\sigma }_{DC}(T)={\sigma }_0\exp \left[-{(T_0/T)}^{1/4}\right]$$

(4)

Here, T₀ is known as a characteristic temperature coefficient which represents a measure of the degree of disorder [28] and may be expressed as:

$$T_0=\frac{\lambda a^3}{kN(E_F)}$$

(5)

where the constant, λ ≅ 16.6, is dimensionless [28], α ≅ 10⁹ m⁻¹ signifies the degree of localization whilst N(E_F) stands for the localized states density at the Fermi level E_F [22,33]. From Equations (2)–(5), after some calculations, the following relation for the N(E_F) is expressed:

$$N(E_F)=\frac{\lambda {(akT)}^3}{{(4E_{A,cond})}^4}$$

(6)

Using the values obtained for E_A,cond (T) from Figure 7 and Equation (6), we computed N(E_F) in the temperature range of 28–120 °C at a low frequency (f = 1 kHz). For the studied samples, the following values were obtained: N(E_F)S1 = 1.1588, 1018 cm⁻³ eV-1; N(E_F)S2 = 1.2888, 1018 cm⁻³ eV-1; and N(E_F)S3 = 1.231, 1018 cm⁻³ eV-1. These results show that the density of states at the Fermi level N(E_F) does not depend on the temperature, remaining constant over the whole investigated domain, as shown recently for other oxide materials (Fe-P or Cu-Mn type) in [34,35]. Also, the N(E_F) for LMO is lower than the values of N(E_F) obtained for LMO:Ga and LMO:Y which contain rare earth ions (Ga and Y, respectively) in the investigated temperature range. This result is in agreement with the fact that the thermal activation energy of electrical conduction, E_A,cond, of the doped materials is smaller than the E_A,cond of the LMO sample (see Figure 7). As a result, we advance the statement that the reduction in E_A,cond is related to the increase in the density of localized states at Fermi level N(E_F), and to support this, we determine two other parameters of the VRH model, i.e., the hopping energy, W, and the hopping distance, R, by means of the following relations [28,33]:

$$R={\left(\frac{9}{8\alpha kTN(E_F)}\right)}^{1/4}$$

(7)

$$W=\frac{3}{4\pi R^{3}N(E_F)}$$

(8)

Making use of the values of $N(E_F)$ in Equation (7), we have computed the hopping distance, R, in every sample and at each measurement temperature within the range of 28–120 °C. With the hopping distance R thus computed, from Equation (8), the hopping energy W was determined for each studied sample. The temperature dependencies of the Mott parameters R and W are shown in Figure 8.

One can see from Figure 8 that increasing the temperature leads to reductions in the hopping distance (Figure 8a) and to increase in the hopping energy (Figure 8b) for all investigated samples. Also, both R and W for the LMO sample are higher than for the LMO:Ga and LMO:Y samples at all the temperatures. This finding aligns with our earlier statement regarding the increase in the density of localized states at the Fermi level, N(E_F), of the modified samples which contain rare earth ions (Ga and Y, respectively) due to the decrease in R and W in these samples compared to the R and W of the LMO sample, thus causing a decrease in the thermal activation energy of conduction, E_A,cond, of Ga- and Y-doped LaMnO₃ relative to the E_A,cond of undoped LaMnO₃.

After these encouraging outcomes for lanthanum manganite compounds, it can be affirmed that further investigations are warranted to ascertain their viability as candidates for energy conversion, such as metal–air batteries or fuel cell electrodes, due to their distinct physical and electronic attributes. The obtained results, namely that the electrical and structural properties of the synthesized Ga- or Y-doped LaMnO₃ ceramic powders can be changed by design and temperature variations, highlight the potential of these materials for utilization in thermo-electric devices and sensor applications.

4. Conclusions

This paper reports on the structural properties and some thermal and electrical characteristics of undoped and Ga- or Y-doped lanthanum manganite, which were synthesized through the sol–gel method, in the R-3c space group (pristine LaMNO₃) and Pm-3m space group (1% Y- and Ga-doped LaMnO₃). The results of the DC electrical conductivity analysis show that the presence of the Ga or Y ions in the LMO perovskite material increases the σ_DC as compared to the σ_DC of the LMO perovskite sample due to the increasing drift mobility of the charge carriers from the samples, in agreement with Mott’s variable-range hopping model. Also, the thermal activation energy of the electrical conduction process, E_A,cond, increases linearly with the increase in temperature for all three samples, but the presence of Ga or Y dopants in the LMO perovskite sample leads to a decrease in the E_A,cond (corresponding to the LMO:Ga and LMO:Y samples) compared to pristine sample (LMO). The following parameters of the VRH model of all samples were determined: the hopping energy of the charge carriers, W; the hopping distance, R; and the density of localized states near the Fermi level, N(E_F). The findings indicate that the N(E_F) stays a constant value throughout the entire temperature range examined, but the presence of Ga or Y rare earth ions in the LMO material leads to a decrease in N(E_F) in these sample compared to the N(E_F) from the LMO sample. At the same time, R decreases and W increases with the increase in temperature for all samples. Also, the presence of Ga or Y ions in the LMO material leads to a decrease in both R and W at constant temperature in correlation with the decrease in the E_A,cond, of doped samples compared to the E_A,cond, of the undoped sample.