Introduction

The discovery of intrinsic colossal magnetoresistance (MR) in hole-doped perovskites with a formula, La1−xAxMnO3 (A = Ca, Sr, and Ba) has established their potential applications in memory devices and magnetic sensors1,2,3,4,5,6,7. However, practical application of the colossal MR is limited by the high magnetic field requirement. Therefore, it is very important to achieve high MR at a low magnetic field. Many reported works have suggested that the regions of structural disorder generated during the film deposition, grain boundaries in short, are the dominant factors in achieving the enhanced low-field magnetoresistance (LFMR). Also introducing a second phase, especially the insulating phase, into La1−xAxMnO3 films has been considered an effective approach for engineering the grain boundaries8,9,10,11,12,13,14,15,16,17,18,19,20,21,22. For example, LCMO:ZrO2 (LCMO: La0.67Ca0.33MnO3) composite films produced by the sol-gel method achieved a MR value of −31% at 77 K under 0.1 T21. Also, −25% of MR at 93 K under 1.15 T was observed in LCMO:15% V2O5 composite fabricated by a two-step solid-state reaction22 and a pronounced LFMR of −23.9% at 10 K in a field of 0.5 T was achieved in LSMO:ZnO (LSMO: La0.67Sr0.33MnO3) composite film14. Finally, a tunable and enhanced LFMR of −30% at 154 K under 0.1 T was observed in a LSMO:ZnO nanocomposite film grown by pulsed-laser deposition (PLD). Most of these reported composite films were, however, prepared by either complicated multiple-steps or costly high-vacuum processes.

Solution deposition method has advantages of low cost, easy setup, and coating of large area as it does not require vacuum-based expensive capital equipment23. Here we report self-assembled LCMO:NiO and LCMO:Co3O4 composite films epitaxially grown by polymer-assisted deposition (PAD) technique which demonstrate enhanced LFMR. Facile and cost-effective PAD method utilizes metal salts dissolved in an aqueous polymer solution24,25,26,27,28,29. Amount of metal cations bound to the polymer and the solution viscosity are directly related to amount of the polymer used. Thus, desired stoichiometry ratio could easily be controlled by mixing different metal-polymer precursor solutions with corresponding metal molar ratios. Thickness of the film could also be controlled by adjusting concentration of the solution or spin-coating rate. The properties of NiO or Co3O4 added LCMO composite films were investigated in both electronic and magnetic perspectives to assess LFMR effects. Under a magnetic field of 3 T, MR values of −44.6% at 255 K for LCMO, −59.1% at 180 K for LCMO:10%-NiO, and −52.7% at 172 K for LCMO:10%-Co3O4 were achieved, respectively. This revealed that the composite films possess high LFMR and their effect was enhanced even further with the addition of 10%-NiO or 10%-Co3O4 insulating phases of the grain boundaries.

Results

Lattice Parameters of Self-epixatially Grown LCMO

Lattice parameters and epitaxial growth were analyzed by in-plane and out-of-plane X-ray diffraction (XRD). XRD data of LCMO:Co3O4 and LCMO:NiO films grown on LaAlO3 (LAO) substrate are shown in Fig. 1. θ–2θ scans in Fig. 1a confirm the introduction of NiO (002) and Co3O4 (004) along with the (002) peaks for LCMO and LAO. The appearance of (00l) only series peaks of the films indicates that the films are highly textured along the c-axis, which is perpendicular to the substrate surface. It is noted that additional peaks appear from the single crystal substrates, apparently the substrates contain impurities. The similar peaks from LAO substrates were found in other reference as well30. The out-of-plane lattice parameters of LCMO:10%- and 30%-Co3O4 were calculated from the LCMO (002) peak to be 3.898 Å and 3.909 Å, respectively, while the out-of-plane lattice parameters of LCMO:10%- and 30%-NiO film were 3.897 Å and 3.891 Å, respectively. As expected, all the values were larger than that of the bulk LCMO (a = 3.858 Å for a pseudocubic perovskite unit cell). This elongation along the c-axis can be attributed to relatively larger lattice parameters of NiO (cubic structure with a = 4.177 Å) and Co3O4 (inverse spinel with a = 8.084 Å). According to the Poisson relationship, the in-plane lattice parameters are compressed when the out-of-plane parameters are expanded. Therefore, we know that the magnitude of the in-plane lattice parameter is in the following order, a (30%-Co3O4) <a (10%-Co3O4) ≈ a (10%-NiO) <a (30%-NiO). Figure 1b and c display the φ scans on reflections of LAO {101}, LCMO {101}, Co3O4 {202}, and NiO {101} of these composite films. The epitaxial growth of LCMO:Co3O4 or LCMO:NiO on LAO could be deduced from these aligned peaks. The {101} peaks of NiO seem to overlap the {101} peaks of LCMO, because the lattice parameter of NiO is close to that of LCMO. The heteroepitaxial relationships between the films (LCMO:Co3O4 and LCMO:NiO composite) and the substrate can be described as (001)LCMO//(002)Co3O4//(001)LAO, [101]LCMO//[202]Co3O4//[101]LAO, and (001)LCMO//(001)NiO//(001)LAO, [101]LCMO//[101]NiO//[101]LAO. Such epitaxial relationships are in accordance with the basal plane lattice parameters of LCMO (a = 3.858 Å), NiO (a = 4.177 Å), Co3O4 (a = 8.084 Å), and LAO (a = 3.79 Å). The lattice mismatch was calculated to be 1.8% between LCMO and LAO; 6.6% between Co3O4 and LAO, while 4.7% between Co3O4 and LCMO; 10.2% between NiO and LAO, and 8.2% between NiO and LCMO. Relatively low lattice mismatch enabled the epitaxial growths of LCMO and Co3O4, LCMO and NiO, LCMO:Co3O4 and LCMO:NiO composites on LAO.

Figure 1
figure 1

 XRD patterns of (a) θ-2θ scan of LCMO:30%-NiO, LCMO:10%-NiO, LCMO:30%-Co3O4 and LCMO:10%-Co3O4 films grown on LAO substrate with additional peaks ¨ from substrates; (b) ϕ-scans of LCMO:30%-NiO from (101) reflections of LCMO, NiO, and LAO; and (c) ϕ-scans LCMO:30%-Co3O4 from (101) reflections of LCMO and LAO, and (202) reflection of Co3O4.

Full size image

Morphology Investigations of the LCMO Composites

Morphology of the films was analyzed using atomic force microscopy (AFM), as shown in Fig. 2. Both the LCMO:30%-NiO and LCMO:30%-Co3O4 composite films showed the uniform surface with the root-mean-square (rms) surface roughness of around 15 nm for LCMO:30%-NiO and 17 nm for LCMO:30%-Co3O4. Triangular NiO and tetrahedral Co3O4 particles were visible on the continuous LCMO matrix surface. The surface of LCMO:10%-NiO is much smoother with rms of 3 nm, indicating that NiO particles are smaller, likely most particles are embedded in the LCMO matrix since they are not obvious on the surface. However, LCMO:10%-Co3O4 has larger Co3O4 particles with different shapes sitting on the surface and the composite has rms of 16 nm.

Figure 2
figure 2

AFM images of (a) LCMO:30%-NiO; (b) LCMO:30%-Co3O4; (c) LCMO:10%-NiO; and (d) LCMO:10%-Co3O4 films grown on LAO substrate.

Full size image

The microstructures of these composite films were also studied by cross-sectional TEM and high resolution TEM (HRTEM). From Fig. 3a,b, in addition to the substrates, two phases are clearly observed in the films as triangular NiO nanoparticles (50–100 nm) and tetrahedral Co3O4 nanoparticles (100–200 nm) on the surface of LCMO. They are consistent with the morphologies shown in the AFM images. The cross-sectional HRTEM images are displayed in Fig. 3c,d. We can see that the interface between the LCMO phase and the substrates are cleanly divided without intermixing. The interface between NiO and LCMO, as shown in the inset, clearly comfirms the epitaxial grown NiO with LCMO matrix. The corresponding selected area electron diffraction (SAED) patterns further confirm the epitaxial growth of LCMO:NiO and LCMO:Co3O4 on LAO by showing the distinct diffraction dots of LCMO, NiO (or Co3O4), and LAO from the SAED patterns; the epitaxial relationships between the composite films and the LAO substrate are identical to the XRD analysis.

Figure 3
figure 3

Bright-field TEM images of (a) LCMO:30%-NiO and (b) LCMO:30%-Co3O4 films grown on LAO substrates; cross-sectional high resolution TEM images of (c) LCMO:30%-NiO with inset image for NiO epitaxially grown with respect to LCMO matrix; and (d) LCMO:30%-Co3O4 on LAO substrates. The insets are the corresponding SAED patterns for each film (Yellow represents LCMO and LCMO-Co3O4, White represents LAO, and green represent NiO).

Full size image

Electrical Properties of the LCMO Composites

By studying temperature dependent resistivity (ρ) over temperature at different applied magnetic fields, different metal-insulator transition temperatures (TPs) were recorded from the composite films. In Fig. 4(a), a well-defined metal-insulator transition feature is displayed from the LCMO film, with the temperature of maximum resistivity at 270 K for 0 T and 295 K for 3 T. Nevertheless, unlike the single phase LCMO films, the composite films do not show metallic behaviors at lower temperatures. For the LCMO:10%-NiO and LCMO:10%-Co3O4 composite films in Fig. 4b,c, respectively, the transition peaks are much broader and less obvious. The TPs under zero field are 180 K for LCMO:10%-NiO and 172 K for LCMO:10%-Co3O4. Moreover, the peak resistivity of these two composite films (0.59 Ω•cm for LCMO:10%-NiO and 1.07 Ω•cm for LCMO:10%-Co3O4 at zero magnetic field) are much higher than that of the single phase LCMO film (0.0061 Ω•cm at zero magnetic field). These two observations can be attributed to the presence of a large number of the grain boundaries, increased disorder, as well as insulating phase-induced barriers to the electrical transport14,31,32,33. The introduced insulating phase at the grain boundary has been known to obstruct the magnetic spin alignment near the grain boundary region. Therefore, it increased the tunneling barrier height between the neighboring magnetic grains13,34,35.

Figure 4
figure 4

Temperature dependent resistivity (ρ) of (a) LCMO, (b) LCMO:10%-NiO, (c) LCMO:10% Co3O4, (d) LCMO:30%-NiO, and (e) LCMO:30%-Co3O4 films at applied magnetic fields of 0 T and 3 T. (f) Plots of adiabatic expression ln(ρ/T) vs 1/T for LCMO, LCMO:10%-NiO, and LCMO:10%-Co3O4 films at temperatures higher than TP and zero magnetic field, with the best fits. The insets are magnifications of the curves around TP.

Full size image

As for the LCMO:30%-NiO and LCMO:30%-Co3O4 composite films shown in Fig. 4d,e, respectively, within the measured temperature range, LCMO:30%-NiO shows a semiconductor behavior without metal-insulator transition, whereas for LCMO:30%-Co3O4, the film manifests an interesting behavior. There is a clear metal-insulator transition at the TP of 265 K in comparison to the other composite films; the resistivity increases sharply with the further temperature decrease. Also, the resistivity increases for the LCMO:30%-Co3O4 composite film in comparison to that of the LCMO:10%-Co3O4 composite film. This is because the metal-insulator transition was suppressed by the expansion of in-plane lattice parameter in LCMO14,22.

As mentioned above, the LCMO:30%-NiO composite film has the largest in-plane lattice constant and this led to the disappearance of metal-insulator transition. On the other hand, for the LCMO:30%-Co3O4 composite film, it has the smallest in-plane lattice constant. Therefore, even though the secondary phase of Co3O4 suppressed the metal-insulator transition, the compressed in-plane lattice constant enhanced the transition, leading to an enhanced TP as shown in Fig. 4(e). The sharp increase in resistivity at low temperatures can be attributed to the possible localization originating from the large amount of Co3O4 secondary phase36,37. Several previous studies have shown that the behavior of paramagnetic phase in LCMO can be well described by the small-polaron transport37,38. For the charge transport by polarons, the resistivity is given by the equation 1:

where ρa is resistivity coefficient, kB is Boltzmann’s constant, and Ea is activation energy that is related to the polaron binding energy13. Therefore, we can have the expression 2:

Figure 4f shows the plots of ln (ρ/T) vs 1/T and their linear fittings for the LCMO, LCMO:10%-NiO, and LCMO:10%-Co3O4 films at temperatures higher than TP with zero magnetic field. The slope of the linear fitting line is proportional to the activation energy Ea. This indicates the energy barrier height for the spin-dependent electron hopping at the grain boundaries17. Thus, we can patently see that the energy barrier height of the LCMO film increases with the addition of the second phase i.e., NiO and Co3O4. In addition, the ln (ρ/T) vs 1/T curves for the LCMO:10%-NiO, and LCMO:10%-Co3O4 films could not show good linear relation. It might be from the variable range hopping mechanism, low-temperature conduction in strongly disordered systems with localized charge-carrier states39.

Magnetic Properties of the LCMO Composites and enhanced LFMR

Enhancement in LFMR is a unique feature and important advantage of the PAD-produced composite films. Figure 5 compares the temperature dependent MR of single phase LCMO and the LCMO composite films with different percentages of NiO and Co3O4, which were calculated from the resistivity at magnetic fields of 0 T and 3 T using the equation, MR (%) = (ρH − ρ0)/ρ0 × 100%. For the composite films, MR curves have several small maxima at different temperatures, while the single phase LCMO film shows a more simplistic curve. This phenomenon may be due to phase inhomogeneity of the composite films29. The maximum MR values increased for the films with the 10% addition of either NiO or Co3O4: −44.6% at 255 K for LCMO, −59.1% at 180 K for LCMO:10%-NiO, and −52.7% at 172 K for LCMO:10%-Co3O4. These MR values are similar to those of other LCMO-based composite films prepared by PLD technique, but the temperature dependent resistivity behavior is different37,40,41,42. For the LCMO:30%-NiO film, the MR value was slightly lower than that of the single phase LCMO film, due to its larger expansion of the in-plane lattice parameter. Whereas, the LCMO:30%-Co3O4 film delivered a comparable MR value to that of the single phase LCMO film. It is worth noting that the TP of LCMO:30%-Co3O4 film was the highest among the four composite samples.

Figure 5
figure 5

The calculated temperature dependent MR of (a) LCMO, LCMO:10%-NiO, and LCMO:30%-NiO films; (b) LCMO, LCMO:10%-Co3O4, and LCMO:30%-Co3O4 films at the applied magnetic field of 3 T.

Full size image

The nature of the magnetically ordered state was investigated by classical zero-field cooled (ZFC) and field cooled (FC) cycles. Figure 6 presents the temperature-dependent magnetization curves measured under an applied field of 100 Oe for the four composite films. All the curves show a similar trend and the FC curves coincides with the ZFC curves at the high temperatures, while the FC curves differ from the ZFC curves in the lower temperature region37,43. It should be mentioned that Curie temperature (TC) of the LCMO:Co3O4 films increases with the addition of Co3O4, whereas for the LCMO:NiO films, TC decreases with the addition of NiO. The enhanced TC for the LCMO:Co3O4 films can be attributed to the decreased in-plane lattice parameter with the addition of Co3O4. On the contrary, the decreased TC for the LCMO:NiO films should be attributed to the increased in-plane lattice parameter with the addition of NiO. Moreover, the similar TCs for the LCMO:10%-NiO and LCMO:10%-Co3O4 owing to the similar in-plane lattice parameter further suggests that the magnetic properties as well as transport properties are dominantly controlled by the in-plane lattice parameter. In addition, the magnetism at low temperatures is clearly suppressed for the LCMO:Co3O4 composite films compared to that of the LCMO:NiO films, which can be explained by the antiferromagnetic properties of Co3O4 phase.

Figure 6
figure 6

ZFC (closed symbols) and FC (open symbols) of magnetization-temperature (M-T) curves measured at 100 Oe for (a) LCMO:NiO and (b) LCMO:Co3O4 composite films.

Full size image

For a facile comparison, out-of-plane lattice parameter, TP , ρ, MR, and TC are summarized in Table 1. Difference between the LCMO:30%-Co3O4 and LCMO:30%-NiO films, and similarity between the LCMO:10%-NiO and LCMO:10%-Co3O4 films in out-of-plane lattice parameter can be seen. In addition, TP and TC show an opposite tendency to the in-plane lattice parameter. LCMO:10%-NiO and LCMO:10%-Co3O4 films show similar TP and TC because of the similar lattice parameters. Meanwhile, the resistivities are different due to distinct conducting properties of Co3O4 and NiO.

Table 1 Summary of parameters of LCMO:NiO and LCMO:Co3O4 composite films prepared from PAD method.
Full size table

Conclusion

In summary, we successfully demonstrated PAD method to fabricate LCMO:NiO and LCMO:Co3O4 epitaxial self-assembled composite films on LAO substrates. PAD method, which provides an alternative yet simpler approach to the growth of self-assembled composite films, is more cost effective compared to the conventional PLD methods. Moreover, the films produced by the PAD method showed enhancement in the LFMR effect which is crucial in the practical application, as demonstrated by the temperature dependent MR analysis. The maximum absolute MR values were further improved by the addition of NiO or Co3O4 due to introduced insulating phase at the grain boundaries. This is entails dramatic increase in resistivity and decrease in TP with the addition of NiO and Co3O4. The temperature-dependent magnetization curve also revealed that TC showed the same trend as TP . More suppressed magnetism was observed from the LCMO:Co3O4 composite films than the LCMO:NiO films on account of the antiferromagnetic properties of the Co3O4 phase. The solution-processed epitaxial films with the enhanced LFMR effect presented here will bring a breakthrough in the field of hole-doped perovskites and their growth.

Methods

Sample preparation

To prepare the precursor solutions for the nanocomposite films, the individual metal-polymer aqueous solutions were prepared first by dissolving 2 g of corresponding metal salts (La(NO3)3∙6H2O, Ca(OH)2, MnCl2∙H2O, NiCl2, or CoCl2∙6H2O) into the polymer solution, which contained 4 g of polyethylenimine (PEI, 50 wt% in water, branched polymer, average Mn ~60,000 by GPC, average Mw ~750,000 by LS, Aldrich) and 2 g of ethylenediaminetetraacetic acid (EDTA, anhydrous, 99%, Aldrich) in 40 g of de-ionized (DI) water. These metal-polymer solutions were then filtered in an Amicon unit, which is designed to pass materials with molecular weight of less than 30,000 g/mol, to remove the unbound ions and concentrate the solutions. The concentrations of La3+, Ca2+, Mn4+, Co3+, and Ni2+ of the final solutions were determined as 125, 191, 148, 245, and 334 mM, respectively, by inductively coupled plasma atomic emission spectroscopy (ICP-AES). These as-prepared solutions were mixed in desired stoichiometric ratios of the single phase LCMO and composite LCMO:Co3O4 or LCMO:NiO (molar ratios of LCMO:Co3O4 or LCMO:NiO are 0.9:0.1 and 0.7:0.3) thin films. The mixed solutions were spin-coated onto LaAlO3 (001) substrates (http://www.mtixtl.com/) at 2,500 rpm for 30 s. The samples were heated at 550 °C for 2 h in flowing oxygen to remove polymer. Films of about 20 nm thickness were obtained from one spin-coat. It is noted that thicker films can be achieved by multiple spin-coats. For the final step, the films were annealed at 950 °C for 2 h in flowing oxygen to improve the crystallization and get epitaxial thin films.

Sample Characterization

The crystal structure of the films was characterized by X-ray diffraction (XRD). The surface morphology and microstructures were analyzed by atomic force microscopy (AFM) and high resolution transmission electron microscopy (HRTEM). The temperature dependence of resistivity was measured by a Quantum Design Physical Property Measurement System (PPMS) along the film surface using a standard four-probe method, with the magnetic field applied normal to the film surface. Temperature-dependent magnetization was measured by a superconducting quantum interference device (SQUID) magnetometer.

Additional Information

How to cite this article: Zhou, M. et al. Enhancement of Low-field Magnetoresistance in Self-Assembled Epitaxial La0.67Ca0.33MnO3:NiO and La0.67Ca0.33MnO3:Co3O4 Composite Films via Polymer-Assisted Deposition. Sci. Rep. 6, 26390; doi: 10.1038/srep26390 (2016).