ABSTRACT

Layered perovskite manganite ceramics with a nominal chemistry La_1.5Ca_1.5Mn_2−xFe_xO₇ (x = 0.0, 0.05, 0.1, 0.3, 0.5) were prepared using sol-gel method. The manganese ions are highly mixed states of Mn³⁺ and Mn⁴⁺. It is found that the Mn³⁺/Mn⁴⁺ ratio decreases with the Fe doping content increasing. The conductivity and magnetoresistance (MR) were studied. The sample of x = 0.05 shows metal insulator transition (MIT) at 135 K. The MIT peak temperature (T_p) shifts towards higher temperature with increasing applied magnetic field. All the samples can be well fitted to the variable-range hopping (VRH) model. The maximum value of MR (%) [ρ(0) − ρ(H)]/ρ(0) × 100 for x = 0.05 is 34% (105 K, 7 kOe).

1. Introduction

The hole-doped ABO₃-type perovskite manganites (Ln, M)MnO₃ (Ln is lanthanide ion, and M is divalent cation) have attracted much attention owing to the colossal magnetoresistance (CMR) exhibited by them [e.g., 1-3]. Substitution of M²⁺ ions for Ln³⁺ leads to ferromagnetism and also metallic conduction. This has been explained as arising due to double exchange (DE) interaction and electron hopping between Mn³⁺ and Mn⁴⁺ cations (Mn⁴⁺ arising due to replacement of Ln³⁺ by M²⁺) via Mn-O-Mn path [4,5]. In the three dimensional perovskites, the Mn³⁺/ Mn⁴⁺ ratio and the microstructure of the Mn–O network are two key parameters controlling the DE interaction, and magnetic properties. La₃Mn₂O₇ like bi-layer manganites also exhibit magneto-resistive behaviour [6] but these are much less studied. In the present paper, the result of temperature and field dependent resistivity measurements on substituting Fe ion for Mn site in calcium substituted bi-layer system La_1.5Ca_1.5Mn₂O₇.

2. Experimental Details

Polycrystalline samples of La_1.5Ca_1.5Mn_2-xFe_xO₇ (x = 0.0, 0.05, 0.1, 0.3, 0.5) were prepared by sol-gel method. The method consisted of adequate citric acid and stoichiometric amounts of the starting materials La(NO₃)₂·6H₂O, Ca(NO₃)₂·6H₂O, Mn(NO₃)₂·4H₂O and Fe(NO₃)₃·6H₂Owhich were dissolved in distilled water, and the pH was adjusted to about 3 to 4 with ammonia solution. First it was heated at 80˚C until a forming the powder. The powder was pressed into thin plates and calcined in air at 1200˚C for 24 h and 1350˚C for 24 h with intermediate grindings between each heating step. To characterize the samples, X-ray diffraction (XRD) measurements were carried [7]. The electrical resistivity was measured at zero field and 7 kOe field by a standard four-point technique. The magnetoresistance (MR) ratio is defined here as Δρ/ρ(0) = [ρ(0) − ρ(H)]/ρ(0), where ρ(0) and ρ(H) denote resistivity at zero field and an applied field, respecttively.

3. Results and Discussion

Figure 1 shows resistivity versus temperature plots for the La_1.5Ca_1.5Mn_2−xFe_xO₇ (x = 0.0, 0.05, 0.1, 0.3 and 0.5) samples at zero field. It is clear from the plots that resistivity enhances with increasing x. It is due to the Mn³⁺/ Mn⁴⁺ ratio reducing with increasing x. This results in the e_g electrons decreasing, which in turn weakens doubleexchange mechanism.

For the samples of x = 0.0 and 0.05, resistivity first rises with increase in temperature, and then exhibits peak around the metal-insulator transition (MIT) temperature denoted by T_p, with the peak temperature T_p about 150 K and 140 K respectively.

For T > T_p, resistivity decreases with increase of temperature. Concentrating on the behavior of temperature dependent resistivity curve, it is noted that dρ/dT < 0 for T > T_p, which is a characteristic of semiconducting behavior. However for T < T_p, the sample showed metallic character with dρ/dT > 0. For higher values of x, the samples show semiconducting behavior and no metalinsulator transition is observed. Usually, the conduction behavior can be described by the a band-gap model, variable range hopping (VRH) mechanism [8-11], or can be described by the adiabatic nearest neighbor hopping model of small polaron conduction (NSPH) [12-14]. An Arrhenius law is used to account for thermal activation process due to a band gap or a mobility edge. In Mott’s VRH model, the resistivity is expressed in the form of, where ρ₀ is the pre-factor and T₀ represents the characteristic temperature. In the nearest-neighbor hopping model of small polaron conduction (NSPH), the resistivity of the sample is given by, where E is the characteristic energy of small polaron hopping. To understand the nature of electronic conduction in the samples, we tried to fit the resistivity to all the three samples. The increasing size disorder with increasing x produces random spin and Coulombic potential fluctuations. So, the carriers find some potential difference beyond the Mn–O distances to hop at farther distances and hence conduct through the Mott’s type of Variable Range Hopping. The T₀ occurring in the VRH relation can be related to the carrier localization length by the expression kT₀ = 24/ L³pN(E), where k is the Boltzmann constant, L the carrier localization length and N(E) is the density of states [8,15]. The values of the localization length should be comparable to Mn-O distances for the VRH type of conduction, here L = 10^{−10 m [16].}

Figure 2 displays the fitting results using the three different functions, a band-gap model, nearest-neighbor hopping model of small polaron conduction (NSPH) and variable-range hopping (VRH) respectively. It can be seen that the fitting results of VRH are better than the other two models i.e. band-gap model and NSPH. This leads us to conclude that the VRH process dominates the conduction mechanism at T > 170 K in these layered manganites. We show the fitting results of VRH model in figure 3. The fitting parameters T₀, N(E) and chi-square are given in Table 1.

From table 1 one finds that the value of N(E) decreases with the x increasing, which indicates that the capacity to conduct reduces with the Fe ion doping. N(E) is also two orders of magnitude higher in these manganites than those of usual oxide semiconductors. Such a higher value of N(E) also estimated by other researchers [8,17], is due to the higher conductivity of these oxides than those of the usual transition-metal-oxide semiconductors [18] and this large value of N(E) is also an indication of adiabatic hopping behavior of the carriers in these manganites as pointed out by Jung [17]. For the NSPH model, the value of E_P augments with the x increasing. It is based on that the amount of e_g electrons reduces with the Fe ion increasing. This results in the resistivity increasing and accordingly the hopping energy increases.

From figure 4 one finds that for the samples of x = 0.05, the resistivity decreases but T_p shifts to higher temperature with the application of magnetic field. This is due to charge carrier delocalization is induced by the magnetic field, which suppresses the resistivity. The application of the magnetic field, also causes the local ordering of magnetic spin and due to this ordering, the ferromagnetic metallic (FMM) state suppresses the paramagnetic insulating (PMI) state. So, the peak temperature (T_p) shifts to the high temperature regime with application of magnetic field.

We discuss below, about magnetoresistance (MR) of the samples of our present investigation. Figure 4 indicates the thermal variation of MR (ρ/ρ(0) (%) = ρ(0) − ρ(H)/ρ(0) × 100, where ρ(H) = resistance in a field of 7 kOe, ρ(0) = resistance at zero field) of two samples. The maximum value of MR for x = 0.05 is 34% (105 K, 7 kOe). The pronounced MR ratio for x = 0.0 and 0.05 is stable in a wide temperature region below T_p. This low temperature MR effect is popular in many n = 2 layered perovskites [6]. Figure 5 show the field variation of MR at 20 K for x = 0.05. As for the heavily doped samples with x = 0.1, 0.3 and 0.5, they present the other kind of ρ-T curves, which are quite different from those of the lightly doped samples.

4. Conclusions

Single-phase rare-earth manganites La_1.5Ca_1.5Mn_2-xFe_xO₇ (x = 0.0, 0.05, 0.1, 0.3 and 0.5) were prepared to study the effect of doping content on the metal-insulator transition and electrical resistivity. It is found that the ratio of Mn³⁺/Mn⁴⁺ reduces with the x increasing. The metalinsulator transition is observed with the lower values of x (x = 0.0, 0.05) and the T_p shifts to higher temperature with the application of magnetic field. The conductivity also depends on the Fe concentration and shows obvious decrease with the increase of Fe content in the samples. It is due to the e_g electrons decrease which in turn weakens double-exchange mechanism. All the ρ-T curves in T > 170 K are fitted quite well in the VRH mechanism. The density of states N(E) is two orders of magnitude higher than those of many transition-metal-oxide semiconductors. The value of N(E) decreases with the x increase, which indicates that the capacity to conduct reduces with the Fe ion doping.

5. Acknowledgements

MPS thanks to Council of Scientific and Industrial Research (CSIR), New Delhi, for the award of fellowship.

REFERENCES