Magnetic and electrical transport properties of Pb_1-xLa_xTi_1-xMn_xO₃ ceramics

Manganites have been well studied not only for their electronic, magnetic and magnetoresistive properties which are important for many applications, but also for their fundamental sciences such as coupling effects.^1,2 Perovskite manganites exemplified by A-site Ca/Sr substituted LaMnO₃ have drawn considerable attention especially following the discovery of colossal magnetoresistance (CMR) effect.¹ It is accepted that both A- and B-site substitutions can have considerable effects on structures, transport and magnetic properties of manganites. The A-site substitution can induce mixed valences of Mn ions and double exchange between Mn³⁺ and Mn⁴⁺, thus macroscopic ferromagnetic nature, metal-insulator transition, and CMR effect. In addition, the mean radius of A-site cations can affect the conduction bandwidth of e_g electrons of Mn ions and control different electronic phases, thus tune the properties.^3,4 Whereas the B-site substitution, especially diamagnetic ions (for example, Al³⁺, Ti⁴⁺) substation, can tune the transition temperature, transport, and magnetic behaviour (ferromagnetic, spin-glass, etc.) by controlling the magnetic dilution and magnetic interaction.^5–9 

On the other hand, in recent years, multiferroic materials having ferromagnetic/ferroelectric orders simultaneously are of great interests due to their potential applications for new electronic devices as well as the fascinating physics of coupling phenomena.¹⁰ However, theoretical works suggest that ferroelectric and ferromagnetic orders are chemically contradicted, therefore, natural single phase multiferroic materials are rare.^11,12 For actual application, multiferroic composite materials, which are constructed by ferroelectric and magnetic materials with different crystal structures, can have large magneto-electric (ME) effect are of interests. The typical composite material is CoFe₂O₄-BaTiO₃, where CoFe₂O₄ and BaTiO₃ are typical ferrimagnetic and ferroelectric materials respectively but have different crystal structures.¹³ However, it is still interesting to find single-phase multiferroic materials because it may be the better research objects to investigate the intrinsic coupling phenomena. One of the possible ways to design single phase multiferroic materials is to prepare solid solutions of magnetic-ferroelectric materials with the same crystal structure, such solid solutions are actually the A- and B-site co-substitutions of one end member for the other end member.^14–17 However, we note that these reported single phase multiferroic solid solutions mainly focus on the compositions with less magnetic end members and emphasize the ME effects, but less attention has been paid to the materials with rich magnetic end member and the corresponding transport and magnetic properties. This is actually important and interesting for further understanding the magnetic and magnetoresistive behavior because the valence states of local magnetic cations may change with substitution and thus the to form coexisted metal ferromagnetic and insulating non-ferromagnetic regions, which has effects on transport and magnetic properties, similar like the grain boundary engineering based composite materials with low field CMR effect.^18,19

Accordingly, in this paper, (1-x)PbTiO₃-xLaMnO₃ [PTLMO100x, x = 0.20, 0.40, 0.50, 0.60, and 0.80] ceramics in a wide composition range have been prepared and investigated. At room temperature, PbTiO₃ (PTO) is a ferroelectric insulator with Curie temperature of 490°C and has tetragonal crystal structure,²⁰ while LaMnO₃ (LMO) is an A-type antiferromagnetic insulator with Néel temperature of 140 K and has orthorhombic crystal structure.^21–23 It is believed that the introduction of Pb²⁺ in A-site may bring local Mn⁴⁺ thus local Mn³⁺-O²-Mn⁴⁺ ferromagnetic interaction, whereas the introduction of diamagnetic cation of Ti⁴⁺ in B-site may dilute and disorder the magnetic interaction between magnetic cations, both will have effects on transport and magnetic properties.

The PTLMO100x (x = 0.20, 0.40, 0.50, 0.60, and 0.80) ceramics were prepared by solid state reaction methods. PbO, TiO₂, La₂O₃, and MnCO₃ (>99.0%) were chosen as starting raw materials. For each composition, the oxides were weighed according to the stoichiometric formula and ball milled for 24 h in ethanol. The dried slurries were calcined at 700°C for 4 h in covered alumina crucibles. The powders were then ball milled for 24 h, dried, calcined at 900°C for 4 h, ball milled for 24 h and dried in sequence. The dried powders were subsequently pressed into rectangle slices (10 mm × 5 mm) with the thickness of 1∼2 mm under 60 MPa. Sintering was carried out at 1200°C for 4 h in covered alumina crucibles. To reduce the volatility of Pb, the slices were embedded in the corresponding powder during sintering.

The crystal structures of the ceramics were characterized by powder x-ray diffraction (XRD, Rigaku Ultima III) using crushed, sintered samples. The microstructures were recorded by scanning electron microscopy (SEM, FEI Quanta 2000). Because the Mn cations can have variable valence states, which are important for magnetic and transport properties, the valence states of the Mn cations were studied by x-ray photoelectron spectroscopy measurements (XPS, Thermo Scientific K-Alpha). Before the chemical state measurements, the surface of ceramics was sputtered by Ar ions (2 keV) for 60 seconds. The magnetic and transport data were collected by using a superconductor quantum interference device (SQUID, Quantum Design, MPMS XL-7) and a physical property measurements system (PPMS Quantum Design, 2001NUGC). The zero-field-cooling (ZFC) and field-cooling (FC, cooling in 200 Oe field) curves were measured with 200 Oe field. The resistivity was measured as the functions of temperature without and with a magnetic field of 2 T, respectively.

Fig. 1 shows the XRD patterns of all the compositions. Clearly, all compositions have well crystallized into perovskite structure. The PLTM20 has a tetragonal structure with the calculated lattice parameters of a = b = 3.9252 Å, and c = 3.9741 Å. However, with the increasing content of orthorhombic LMO (x), the structure changes to be orthorhombic, as confirmed by the observation that the splitted diffraction peaks tend to merge into one peak. Furthermore, it is noticed that PLTM40, PLTM50, and PLTM60 have almost same peak positions, indicating their close lattice parameters, which are calculated to be a = 5.5466 Å, b = 5.5419 Å, and c = 7.8217 Å. The diffraction peaks of PLTM80 show high-angle shift when compared with other compositions, indicating the decreased lattice parameters which are calculated to be a = 5.5315 Å, b = 5.5251 Å, and c = 7.7978 Å. The LMO induced lattice decrease should be attributed to that LMO has relatively smaller lattice constants than PTO, either in orthorhombic or pseudocubic crystal structures.^20,23 Figs. 2(a)–2(e) show the microstructures of the compositions with x = 0.20, 0.40, 0.50, 0.60, and 0.80, respectively. As can be seen, with increasing x, the grain size tends to decrease. Interestingly, PLTM20, PLTM40 and PLTM50 has relatively dense microstructures than PLTM60 and PLTM80. This can be attributed to the low melting temperature of PbO (∼ 890°C),²⁴ which will is helpful to densify the microstructure by liquid-phase sintering.²⁵ So, it is reasonable decreased PTO content (i.e., large x value) will lead to decreased dense microstructure.

The chemical state of Mn cations for all compositions is investigated by XPS measurements, and the Mn 2p spectra are plotted in Fig. 3. For all compositions there are two main peaks locating near 641 eV and 653 eV, which correspond to the doublet of Mn 2p_3/2 and Mn 2p_1/2 and indicate the existence of Mn³⁺.²⁶ The spin-orbit splitting of the Mn 2p level is about 12 eV, which is consistent with other reports.^26–28 Because the Mn 2p levels of Mn³⁺ and Mn⁴⁺ have very similar band structures,^27,29 it is difficult to observe any independent XPS peak arising from Mn⁴⁺. However, with increasing x to 0.50, 0.60, and 0.80, a satellite peak locating near 647 eV, indicated by arrows, tends to appear. This peak is attributed to the mixed states of Mn³⁺ and Mn⁴⁺, similar results as observed in A-site Sr/Ca substituted LMO materials.^27,28 Accordingly, the valence state of Mn cations is +3 for the PLTM20 and PLTM40, whereas is the mixture of +3 and +4 for the PLTM50, PLTM60, and PLTM80.

Fig. 4(a) shows the magnetization-magnetic field (M-H) curves measured at 10 K of all compositions, and Fig. 4(b) is the locally enlarged loops. The temperature dependent magnetizations of all compositions are plotted in Fig. 5. For the PLTM20 and PLTM40, the S-typed M-H curves are not saturated under 2 T field and the curves have no hysteresis. In addition, the magnetization increases monotonously with decreasing temperature, all these results indicate that superparamagnetic behavior has been detected. By considering that for these two compositions, the magnetic Mn³⁺ cations dilute randomly in paramagnetic insulating matrix to form isolated magnetic region, which may have no interaction with each other because the matrix prevents the magnetic interaction. And this should be responsible for the observed superparamagnetic behavior of these two compositions. Actually, similar phenomena have been reported in both a diluted magnetic semiconductor and a similar system, Sr_2-xLa_xMnTiO₆.^5,30

On the other hand, for the PLTM50 and PLTM60, the M-H curves show obvious hysteresis, although not well-saturated. From the FC-ZFC curves in Fig. 5, it is found that with decreasing temperature from room temperature to about 20 K, both the FC-ZFC curves of the PLTM50 and PLTM60 increase monotonously. With further decreasing temperature, the FC curves tend to saturate and the ZFC curves tend to decrease to form a cusp, i.e., the ZFC-FC curves split. One of the possible reasons for such phenomena is the coexistence of antiferromagnetic and ferromagnetic order.^7,31 By further considering the M-H hysteresis loops (Fig. 4), the coexistence of Mn³⁺ and Mn⁴⁺ (Fig. 3), and the magnetic Mn cations are not locally distributed because magnetic cations are not the diluted end member, it is reasonable that local ferromagnetic clusters have been formed due to the Mn³⁺-O^2--Mn⁴⁺ double exchanges, where as there are still considerable Mn³⁺-O^2--Mn³⁺ antiferromagnetic interactions. In other words, ferromagnetic and antiferromagnetic interactions coexist in the PLTM50 and PLTM60.

For the PLTM80, the ferromagnetic interaction should be dramatically enhanced and the antiferromagnetic interaction weakened because of the following reasons. 1) The increased Mn⁴⁺ cations as confirmed by the XPS result in Fig. 3. 2) The cusp of the ZFC curve is significantly suppressed. 3) The M-H hysteresis loop is well-saturated and the magnetization decreases sharply to near zero in the temperature range around 120 K, which should be the ferromagnetic-paramagnetic phase transition temperature.^8,32,33 However, the existence of antiferromagnetic interaction in this composition cannot be excluded because of the visible cusp of the ZFC curve as well as the transport behavior shown below.

The transport properties of all compositions are shown in Fig. 6(a). The PLTM20 has too large resistivity for transport measurement. The resistivity decreases significantly by nearly 3 orders with increasing x from 0.20 to 0.80 at 300 K. But all the compositions show insulator-like behavior in the whole temperature range, which is due to the existence of nonmagnetic insulating regions (PTO for example). Furthermore, the resistivity of the PLTM80 increases flatly with decreasing temperature, especially when temperature is lower than 120 K, which might be attributed to the significantly enhanced Mn³⁺-O^2--Mn⁴⁺ metallic ferromagnetic interaction. It is interesting to note that the PLTM80 shows ferromagnetic insulator character. This may due to the presence antiferromagnetic interactions as well as nonmagnetic insulating regions (PTO for example) prevent the current from crossing.

Furthermore, the PLTM20, PLTM40, PLTM50, and PLTM60 shows small negative MR (<6%). However, the PLTM80 shows large magnetoresistance effect, the negative MR can reach −41.5% at 10 K with the field of 2 T, as shown in Fig. 6. The negative MR decreases with decreasing temperature and a negative dip near the phase transition temperature (∼120 K) is observed, which is attributed to the phase transition, similar dip have been observed in LaMnO₃/SrMnO₃ superlattice with coexisted ferromagnetic and antiferromagnetic orders.²³ The observed large magnetoresistance might arise from that for this composition, there are non-ferromagnetic insulators locating at the boundaries of the ferromagnetic grains, which can increase the electronic barrier in the grain boundaries, thus leading to the spin-polarized tunneling or spin dependent scattering at the grain boundaries. Actually, the enhanced negative MR by introducing non-magnetic insulating grain boundaries are of the interesting research topics for long time and have been widely observed and discussed.^18,19

In summary, Pb_1-xLa_xTi_1-xMn_xO₃ (x = 0.20, 0.40, 0.50, 0.60, and 0.80) ceramics have been prepared and the structures, chemical states, magnetic and transport properties have been investigated. The Mn cations have the valence state of +3 in the compositions with x = 0.20 and 0.40, which show superparamagnetic behaviour, whereas mixed +3 and +4 in other compositions, which show coexisted ferromagnetic and antiferromagnetic behaviours. All compositions have insulating behaviour. It is interesting to find the composition with x = 0.80 has a large negative magnetoresistance of −41.5% at 10 K with the applied filed of 2 T. The observations are explainable based on the model that the metallic ferromagnetic cluster embedded in insulating non-ferromagnetic matrix.

This work was supported by the National Nature Science Foundation of China (11174127), the Doctoral Fund of Ministry of Education of China (20110091110014), the Nature Science Foundation of Jiangsu Province (BK2009007), and the Priority Academic Program Development of Jiangsu Higher Education Institutions.