Fe-doping and strain effects on structural and magnetotransport properties in La 2 Õ 3 Ca 1 Õ 3 Mn 1 − y Fe y O 3 thin films

The influence of Fe-doping and strain effects on the structural and magnetotransport properties of undoped and lightly doped Fe 1 and 3% at Mn site La2/3Ca1/3MnO3 thin films and bulk powder samples have been studied. Thin films were grown on 100 -SrTiO3 STO and 100 -LaAlO3 LAO single crystal substrates, via high O2 pressure 500 mTorr using dc magnetron sputtering. Conversion electron Mössbauer CEM spectra measured at room temperature in the paramagnetic regime of the Fe-doped samples do not show significant differences in the isomeric shift for the case of the La2/3Ca1/3MnO3 films doped with 1 and 3% iron. The isomeric shift values correspond to the presence of Fe in the 3+ state with octahedral coordination, thus indicating that Fe is incorporated into the structure by substituting Mn. The absence of further states in the spectra indicates that Fe is not involved in forming other additional impurity phases. The x-ray -2 scan showed that all thin films on LAO and STO have single phase and c-axis strong orientation along the growth direction and the Fe doping gives rise to a relaxation of the epitaxial strain. Finally, we have observed that the saturation magnetization, Curie temperature, metal-insulator transition, and magnetoresistance vary nonmonotonically with increased Fe concentration. This behavior can be understood in terms of competing influences from the strain relaxation, which enhances the tendency to order ferromagnetically, and the reduced double exchange, which is detrimental to the ferromagnetic order.


I. INTRODUCTION
The mixed valence perovskite manganites of type La 1−x A x MnO 3 ͑A = divalent ion, such as Ca, Sr, Ba, or Pb͒ have been extensively investigated because of a wide variety of interesting phenomena such as colossal magnetoresistance ͑CMR͒, half-metallicity, and phase separation. 1,24][5] The CMR manifests itself by a large decrease in the resistivity, when an external magnetic field is applied close to the Curie temperature T C ͑Ref.3͒.One main focus for theoretical and experimental research on mixed valence manganites is the potential applications, as well as, understanding the physical mechanism responsible for the large magnetoresistance ͑MR͒.From an applied point of view it is desirable to increase the values of the MR at low applied magnetic field and to shift the MR peak close to room temperature. 1,2][8] However, there are experimental and theoretical evidence that suggest that the DE model alone is insufficient to explain the CMR and the other variety of phenomena observed in these compounds. 9n this model the unpaired e g electron in the high spin configuration of Mn 3+ moves to a neighboring Mn 4+ .This electron transfer is favored if the Mn 4+ receives the electron with its spin aligned parallel to the total spin of Mn 3+ .In this scenario, electron delocalization between Mn ions leads to a lowering of energy and will favor a ferromagnetic ordering of the magnetic moments.For a complete understanding of the magnetic properties it is however important to understand how this double exchange competes with the Mn 3+ -O-Mn 4+ superexchange. 7,8In manganites the doping of a divalent ion, e.g., Ca, onto the La site induces mixed valences of Mn 3+ / Mn 4+ , where the Mn 3+ ions can produce a ferromagnetic double-exchange interaction resulting in the CMR behavior. 1,2An interesting way to modify the physical properties of these materials is to dope on the Mn site of the crucial Mn 3+ -O-Mn 4+ network, which is at the core of double exchange. 10The introduction of other trivalent ͑Al, In͒ or transition metals ͑Fe, Ni, Co, Cr, Cu, etc.͒, which introduce a different electronic configuration to the Mn site, shall lead to effects associated with the electronic configuration mismatch between Mn and the other substituting ions.For example, Blasco et al., Sanchez et al., Ghosh et al., and Rivadulla et  al. have studied the effect of the substitution of Mn with Al, In, Co, and Cr in the magnetoresistive La 2/3 Ca 1/3 MnO 3 perovskite.They have observed that the compounds became magnetically disordered and inhomogeneous, which was accompanied by structural disorder via the formation of oxygen vacancies in the Mn sublattice as a result of the doping in the Mn site.][12][13] Similar to the aforementioned studies, various investigations of electrical, structural, and magnetic properties of polycrystalline, doped manganites have attempted to understand the influence of Fe doping on the magnetotransport behavior.Fe is one of the nearest neighbors of Mn in the periodic table of elements and its ionic radius is close to the Mn ionic radius. 14In addition, Fe belongs to the third subgroup of elements in the periodic table, and thus both Fe and Mn have similar energy level structure and multiple valence; thus Fe is able to substitute for either Mn 3+ or Mn 4+ .Because of this, Fe doping on the Mn site is expected to show inter-esting physical properties and should provide useful information for understanding the physical mechanisms in these systems.Additionally, Mössbauer spectroscopy ͑MS͒ allows microscopic type of measurements, which are particularly relevant and provide information about minute changes in the local electronic, magnetic, and vibrational state of 57 Fe via hyperfine interaction.The most relevant past investigations of polycrystalline, Fe-doped manganites are as follows: Leung et al. studied a polycrystalline sample of La 1−x Pb x Mn 1−y Fe y O 3 and observed that magnetic moments of the Fe ions are mostly antiparallel to the Mn magnetization and reduce the magnetization and Curie temperature through the suppression of double exchange. 15Later, Ahn et al. observed a suppression of ferromagnetism and conductivity in both the ferromagnetic ͑FM͒ and the antiferromagnetic ͑AF͒ phase of La 1−x Ca x MnO 3 by doping Fe on the Mn site.Lattice effects, which often dominate doping effects, are negligible in this case due to the nearly identical size of Mn 3+ and Fe 3+ .Again the suppression of ferromagnetism and conductivity originates from the reduction of double exchange due to the depopulation of hopping electrons by the Fe doping. 16Subsequently, Pissas et al. and Righi et al. studied lightly Fe-doped polycrystalline La 1−x Ca x MnO 3 compounds and observed behavior consistent with variable sized FM clusters and AF coupling of the Fe atoms, resulting in a large influence of the Fe doping on the transport properties as well as the magnetic behavior. 17,18Later, Cai et al. found spinglass behavior with a spin freezing temperature of 42 K close to the metal-insulator transition and CMR effect in the perovskite La 0.67 Ca 0.33 Mn 0.9 Fe 0.1 O 3 .These results were ascribed to the formation of FM and AF clusters and the competition between them with the introduction of Fe 3+ ions, which are not involved in the DE. 19Afterwards, Ogale et al. examined La 0.75 Ca 0.25 Mn 1−y Fe y O 3 for y values from 0 to 0.05 and suggested that transport in this system involves quasiparticle excitation in association with the DE process and it undergoes localization-delocalization-type transition at a critical dopant concentration. 20Simopoulos et al. have shown that the Fe moments are antiferromagnetically coupled to the ferromagnetic Mn host, when the Fe substitutes for Mn as Fe 3+ in La 2/3 Ca 1/3 MnO 3 doped with Fe. 21Hannoyer et al. and Yusuf et al. also observed the existence of FM domains in the grain boundary regions of La 1−x Ca x Mn 1−y Fe y O 3 polycrystalline samples. 22,23emarkably, although the properties of the Fe-doped manganites La 1−x Ca x MnO 3 in polycrystalline powder and ceramic samples have been well studied, only very limited studies have been performed on thin films of Fe-doped La 2/3 Ca 1/3 MnO 3 .Huang et al. reported an AF arrangement of the Fe and Mn ions over the whole Fe-doping range and a canted spin structure at y Ն 0.12 in La 0.7 Sr 0.3 Mn 1−y Fe y O 3 thin films prepared by pulsed laser deposition.They suggested that increased MR arises from the suppression of enhanced magnetic and polaron scattering under the application of an external field. 24Finally, Bathe et al. have explored the effect of 57 Fe ion implantation on the magnetotransport of epitaxial La 0.67 Ca 0.33 MnO 3 thin films.An increase in the resistance was observed which was attributed to the defects and disorder produced on the surface of the films by Fe ion implantation. 25 this study, we investigated Fe doping onto the magnetic Mn lattice site and examined its implications for the structural, magnetic and magnetotransport properties in powders as well as in thin films.Furthermore we studied stress effects due to different thicknesses and substrates.We employed the 57 Fe nucleus as a microscopic probe to detect the local behavior of magnetic ions in a series of La 2/3 Ca 1/3 Mn 1−y Fe y O 3 samples, with y =0 ͑LCMO͒, 0.01 ͑LCMFO-1%͒, and 0.03 ͑LCMFO-3%͒.It turns out that the effects of Fe doping and epitaxial stress have competing impacts on structural and magnetotransport properties.The doping relaxes the epitaxial strain efficiently, which is beneficial to FM order, and at the same time inhibits FM order by suppressing double exchange.

II. EXPERIMENTS
The films of La 2/3 Ca 1/3 Mn 1−y Fe y O 3 ͑y = 0, 0.01, and 0.03͒ used in this study have been grown on SrTiO 3 ͑100͒ and LaAlO 3 ͑100͒ single-crystal substrates using dc magnetron sputtering.The La 2/3 Ca 1/3 MnO 3 ͑LCMO͒, La 2/3 Ca 1/3 Mn 0.99 Fe 0.01 O 3 ͑LCMFO-1%͒, and La 2/3 Ca 1/3 Mn 0.97 Fe 0.03 O 3 ͑LCMFO-3%͒ powders for targets were prepared by a standard solid state reaction technique, using high purity ͑Ͼ99.9%͒oxide powder of La 2 O 3 , MnO 2 , CaCO 3 , and Fe 2 O 3 96.05%enriched with 57 Fe, in order to enhance the Mössbauer signal at the lowest doping level.The powders of these oxides were mixed and heated in air at 1200 °C for several days.All deposition parameters were kept constant for all the films, except the deposition time.During deposition the substrate temperature was kept at 850 °C, the high purity oxygen pressure was 500 mTorr, the substrate-target distance was 40 mm and the deposition power was 40 W.After deposition, the films were cooled down to room temperature in a high-purity oxygen atmosphere at 750 Torr pressure.A series of films LCMO, LCMFO-1% and LCMFO-3% were grown with thicknesses ranging between 23-80 nm.
The structure of the powder for the targets and the deposited films were characterized by high-angle x-ray diffraction ͑XRD͒ using a Philips X'PERT PRO diffractometer with Cu K␣ radiation.The film thicknesses were calculated from grazing incidence x-ray reflectivity measurements.The composition and morphology were analyzed using scanning electron microscopy ͑SEM͒ and energy dispersive spectroscopy ͑EDX͒.
The local environment of 57 Fe atoms in LCMFO-1% and LCMFO-3% target powders and films were investigated by Mössbauer effect ͑ME͒ measurements.ME spectra on powder samples were recorded in transmission geometry using a conventional spectrometer with a 57 Co͑Rh͒ source of 25 mCi strength moving in a constant acceleration mode at room temperature.The calibration of the spectrometer was made at room temperature using a thin ␣-Fe foil as the absorber, and the spectra were fitted by using the computer program MOSFT.26 Conversion electron Mössbauer spectroscopy ͑CEMS͒ was used to get information about Fe behavior in the films.CEM spectra were recorded at room temperature using a flowing gas ͑95% He, 5% CH 4 ͒ proportional counter.
The evaluation of CEM spectra was performed by using MOSSWINN 27 programs.
Transport measurements in the films as a function of temperature ͑5 Յ T Յ 300 K͒ and magnetic field ͑from −5 − 5 T͒ have been carried out in a 6 T transverse superconducting magnet system using a dc four-probe method.Electrical contacts were attached to unpatterned films via indium soldering with a 2 mm separation between contacts.Finally, magnetic measurements were carried out with a Quantum Design superconducting quantum interference device ͑SQUID͒ magnetometer over a temperature range of 5-300 K.The magnetization vs temperature M͑T͒ in zero field cooling ͑ZFC͒ was measured with an applied external magnetic field H of 2 kOe.The M͑H͒ curves were measured from −1 − 1 T at various temperatures in order to investigate the temperature evolution of the coercive fields.In both measurements, the applied magnetic field was parallel to the film surface.

III. RESULTS AND DISCUSSION
The XRD pattern of the LCMO, LCMFO-1%, and LCMFO-3% powders and films are shown in Figs.1͑a͒ and 1͑b͒ and were collected by scanning over angular range 20 Յ 2 Յ 80°with steps of 0.01°.The measurements reveal that the three XRD patterns of the La 2/3 Ca 1/3 Mn 1−y Fe y O 3 powders are single-phase perovskite without detectable secondary phase or impurity, as shown in Fig. 1͑a͒.The refinement of the XRD patterns for the powder samples was carried out by using the MAUD program 28 assuming for all samples the perovskite orthorhombic structure and using the Pbnm ͑62͒ space group.The fitted lattice parameters are shown in Table I.The b and c axis of the doped samples are just a little bit smaller than the values for undoped LCMO.Both the lattice parameters and the cell volumes decreased monotonically with Fe doping.These results indicate that the replacement of Mn ions by Fe ions does not have much influence on its lattice because of the identical ion size of Fe 3+ and Mn 3+ .A similar behavior was observed by Cai et al. for La 0.67 Ca 0.33 Mn 0.9 Fe 0.1 O 3 ͑Ref.19͒. Figure 1͑b͒ shows typical XRD patterns of LCMO, LCMFO-1%, and LCMFO-3% films grown on LaAlO 3 ͑LAO͒ substrates.XRD patterns on SrTiO 3 ͑STO͒ are not shown, but are similar.The x-ray -2 scan showed that all thin films on LAO and STO are single phase and are c-axis oriented along the growth direction.All reflection peaks corresponding to the thin films on LAO and STO were fitted using Gaussian and Lorentzian functions as shown in the inset of Fig. 1͑b͒.From the fits of the peak positions we calculated the lattice constants of the − a film ͒ / a Subs ͔ ‫ء‬ 100%, where a Subs are the values for the substrate used, a LAO = 3.78 Å, a STO = 3.9 Å, and a films are lattice parameters of the perovskite cubic unit cell, which were calculated taking into account that the crystalline structure of LCMO, LCMFO-1%, and LCMFO-3% thin films on LAO and STO, are of orthorhombic type, with lattice parameters a ϳ 2 1/2 a film , b ϳ 2 1/2 a film , and c =2a film .

Films
Thickness ͑nm͒    II show that with increasing iron concentration, the distance between reflection planes is shifted to lower values, which is qualitatively consistent with the behavior observed for the powder samples.Also, the calculated lattice parameters of the LCMO, LCMFO-1%, and LCMFO-3% thin films grown on the STO and LAO substrates ͑which have different lattice parameters͒ reveal the existence of a competition between the epitaxial strain and iron doping effects.Figure 2 shows that the lattice parameters of the films relax with the increasing iron concentration for films grown on both STO and LAO.The strain effects are more notable for the thinnest films.The strain and iron doping effects can be quantified better by calculating the lattice mismatch ␦ of the films along the interface with re-gard to the lattice parameters of the substrate ͑see ␦ values in Table II͒.Positive values of ␦ correspond to tensile strain at the interface, where the unit cell stretches along the plane of the film and is compressed along the growth direction out of the plane.In contrast, negative values describe compressive strain, where the cell stretches along the growth direction and is compressed in the plane of the film. 29For the case of the films grown on STO, the mismatch decreases, and therefore the lattice parameter relaxes when the iron concentration increases.This indicates that the iron concentration leads to the decrease in the strain-induced distortion of the unit cell.
A similar behavior occurs in the films grown on LAO with different thicknesses.Notice that for 1% and 3% Fe doping, the c-lattice constant is almost independent of the substrate choice, while there is an appreciable difference for the undoped films.This indicates that while there is significant epitaxial strain for the undoped samples, especially when they are grown on LAO, the epitaxial strain is negligible for the doped films.Therefore the incorporation of Fe gives rise to defects, which effectively relaxes the strain.
Figure 3 shows the Mössbauer spectra of the LCMFO-1% and LCMFO-3% powder samples in the paramagnetic state at room temperature.In both cases the spectra were very similar and consist of a single absorption line typical of paramagnetic behavior.These spectra were fitted with a quadru-pole doublet having the following hyperfine parameters: isomer shift IS = 0.32͑3͒ mm/ s, quadrupolar splitting QS = 0.21͑6͒ mm/ s, linewidth w = 0.24͑1͒ mm/ s for LCMFO-1% sample and IS = 0.338͑1͒, QS = 0.281͑4͒ mm/ s, and w = 0.406͑6͒ mm/ s for LCMFO-3%.The isomer shifts are typical of Fe 3+ ions in octahedral coordination, while the small quadrupole splitting indicates a slight distortion of the octahedron. 21,30n order to analyze the local behavior of iron in the thin film samples, CEM spectra were collected on LCMFO-1% and LCMFO-3% grown on STO.Experimental data of CEM spectra and their respective fits are shown in Fig. 4. The spectrum of LCMFO-1% films was fitted using a singlet with  a Lorentzian shape as shown in Fig. 4͑a͒.The relatively noisy baseline is due to the low 57 Fe concentration in this weakly doped film.The spectrum shape is typical of a paramagnetic sample, which is the expected behavior at room temperature.From the fits, the hyperfine parameter values are determined to be IS = 0.345͑3͒ mm/ s and w = 0.525͑6͒ mm/ s.For the LCMFO-3% films, the CEM spectra are similar to LCMFO-1% films, as Fig. 4͑b͒ shows.In contrast to the samples doped with 1% of Fe, the CEM spectrum for 3% iron doping is better resolved and has less noise for the baseline.Again, the experimental data were fitted using a singlet resulting in hyperfine parameters IS = 0.354͑7͒ mm/ s and w = 0.537͑4͒ mm/ s.As for the powder samples the IS values are ϳ0.35mm/ s, which is consistent with Fe 3+ , without coexistence of another charge state.However, for the films the site surrounding the iron was found undistorted, in contrast to the observed quadrupolar splitting for the powder samples.This behavior is probably attributed to the epitaxy, which may lead to a better incorporation of the iron ions in the manganite structure.FIG. 10.Temperature dependence of the without external applied magnetic field for LCMO, LCMFO-1%, and LCMFO-3% thin films grown on ͑a͒ STO and ͑b͒ LAO.To enable a better comparison, all resistance data were normalized to the 290 K values.Magnetoresistance ⌬R / R 5T as a function of temperature for LCMO, LCMFO-1%, and LCMFO-3% thin films grown on ͑c͒ STO and ͑d͒ LAO substrates.Inset: resistance as a function of temperature at H = 0 T and 5 T magnetic external applied field for LCMFO-1% thin films on STO.
Furthermore the CEM spectra showed no evidence of other oxide phases which is consistent with the XRD measurements.
The temperature dependencies of the magnetization of the powder samples LCMO, LCMFO-1%, and LCMFO-3% are shown in Fig. 5 and were measured with increasing temperature and an applied field of 2 kOe after cooling from room temperature to 5 K in zero field.The ferromagneticparamagnetic phase transition is clearly seen and the transition is very sharp for the undoped sample, while it is broader for the doped samples possibly due to structural disorder in the Mn lattice.A similar behavior was found in samples of La 2/3 Ca 1/3 Mn 1−x In x O 3 with low In concentration ͑x Յ 0.03͒ ͑Ref.10͒.The Curie temperature T C is determined by the inflection point of the magnetization curve vs T. For the undoped sample LCMO we find T C ϳ 250.7 K and the saturation magnetization M S ϳ 93.7 emu/ g.The introduction of Fe doping lowers both T C and M S by ϳ4.5 K and 5.2 emu/ g for the 1% Fe-doped sample and ϳ7.8 K and 7.0 emu/g for the 3% Fe-doped sample ͑see Fig. 5 and inset͒.Notice that the relative reduction of the saturation magnetization is slightly larger than what would be expected from the fraction of Mn replaced by Fe.This indicates that not only are the Fe moments oriented antiparallel to the overall ferromagnetically ordered Mn moments, but also the Fe destroys the ferromagnetic order locally.
The temperature dependence of the magnetization of LCMO, LCMFO-1%, and LCMFO-3% thin films is shown in Fig. 6͑a͒ for STO substrates and in Fig. 6͑b͒ for LAO substrates.For all films the T C values are lower than for the corresponding bulk samples.Remarkably the T C depends nonmonotonically on the Fe concentration as is also shown in Fig. 7.This behavior of the critical temperature is caused by the competition between the strain relaxation and reduced double exchange due to the Fe doping.Interestingly, the different tensile ͑STO͒ and compressive ͑LAO͒ strain gives rise to very different T C values for the undoped films, while for the doped films the T C is nearly independent of the choice of substrate.This indicates that the Fe doping allows for the strain to be completely relaxed even for the thinnest films.
The hysteresis loops measured at 5 K in LCMO, LCMFO-1%, and LCMFO-3% thin films on STO and LAO are shown in Fig. 8.In both cases, the coercivities increased with Fe doping as is shown in Fig. 9͑a͒, which is consistent with an increased defect density upon doping.These defects may result in increased pinning of domain walls.The Fe-doping dependence of M S is shown in Fig. 9͑b͒ for LCMO, LCMFO-1%, and LCMFO-3% thin films on STO and LAO.For films with comparable thicknesses the M S values on LAO are larger than for STO.In contrast to the doping dependence of T C , M S still depends on the substrate choice for 1% doping.This probably indicates that for the 1% doping there is still a significant fraction of the film strained close to the interface.But at the 3% doping the epitaxial strain is again completely negligible.
The temperature dependence of the electrical resistance R measured under zero applied magnetic field for the LCMO, LCMFO-1%, and LCMFO-3% thin films grown on STO and LAO are shown in Figs.10͑a͒ and 10͑b͒.For a better comparison of all curves, they are normalized to the resistance at 290 K.All films show a peak in the resistance, which is associated with a metal-insulator transition.
Upon applying a 5 T field T MI is significantly decreased for all samples as is shown in Table III.This results in magnetoresistances MR = ⌬R / R 5T = ͑R 5T -R 0T ͒ / R 5T , which are shown in Figs.10͑c͒ and 10͑d͒.It is interesting to note that the peak magnitude of MR increases with decreasing T MI and thus the maximum MR shows a nonmonotonic doping dependence mirroring the doping dependence of T MI ; see Figs. 11͑a͒ and 11͑b͒.The metal-insulator transition temperature T MI is for all films consistent with T C obtained from magnetization measurements, i.e., compare Figs.7 and 11͑a͒ and also see Table III.This means that T MI also shows a nonmonotonic doping dependence reflecting the interplay between strain relaxation and double-exchange suppression.This tendency of MR has also been observed in studies re- 12 Surprisingly, the maximum MR is still very sensitive to the substrate, even for the highest doping, although all the magnetic properties are essentially identical.As can be seen in Figs.10͑c͒, 10͑d͒, and 11͑b͒ the maximum MR for the 3% sample is almost twice as high for samples grown on LAO compared to samples grown on STO.This may indicate that the magnetoresistance could be extremely sensitive to even very small differences in the remaining strain.
Finally, in Fig. 12 we show the relative resistance measurements R͑H͒ / R͑0͒ as a function of external magnetic field measured at the temperatures corresponding to the peaks observed in Fig. 10.We note that the width of the magnetore-sistance curves reflect again the nonmonotonic doping dependence and that the width decreases with decreasing T MI .

IV. CONCLUSION
We have grown La 2/3 Ca 1/3 Mn 1−y Fe y O 3 ͑y = 0, 0.01, and 0.03͒ thin films on SrTiO 3 and LaAlO 3 single-crystal substrates using dc magnetron sputtering.CEMS measurements showed that Fe is incorporated only in the 3+ state and that there are no significant impurity phases containing Fe.The lattice mismatch between the films and the substrate relaxes in the Fe-doped samples and in turn greatly affects the magnetization and magnetotransport properties.We have observed that the saturation magnetization, Curie temperature, metal-insulator transition, and magnetoresistance vary nonmonotonically with increasing Fe concentration.This behavior can be understood in terms of competing influences from the strain relaxation, which enhance the tendency to order ferromagnetically, and the suppression of double exchange, which is detrimental to the ferromagnetic order.

)FIG. 1 .
FIG. 1. ͑a͒The XRD spectra of LCMFO-3%, LCMFO-1%, and LCMO powder samples.The open circles are the experimental data, and continuous lines are the fits.For LCMO the Miller indices to the corresponding planes are shown.͑b͒ The XRD spectra of LCMFO-3%, LCMFO-1%, and LCMO thin films grown on LaAlO 3 ͑LAO͒.Asterisks indicate the manganite peaks and the circles indicate enlarged regions with their respective fits.The fits of experimental data were carried out using Lorentzian and Gaussian functions.

FIG.
FIG. TransmissionMössbauer spectra of ͑a͒ LCMFO-1%, ͑b͒ LCMFO-3% powder samples at room temperature in the paramagnetic regime.The filled circles are the experimental data and continuous lines are the fits.

TABLE II .
Thickness t, distance between reflection planes d, lattice constants c, and lattice mismatch ␦ defined as ͓͑a Subs
unit cell for the films.TableIIsummarizes the lattice parameters for LCMO, LCMFO-1%, and LCMFO-3% films on STO and LAO.The results from Table

TABLE III .
Metal-insulator transition temperature T MI for H = 0 and 5 T, thickness and ferromagnetic-paramagnetic transition temperature T C for LCMO, LCMFO-1%, and LCMFO-3% thin films on STO and LAO.͑a͒ Fe-doping dependence of metal-insulator transition temperature T MI and ͑b͒ MR maximum values, for thin films grown on STO and LAO for different thicknesses between 23 and 80 nm.The triangle and square symbols represent the thicker and thinner samples grown on LAO, respectively, while open circles connected with dashed lines correspond to the samples on STO.ported on metal transition elements ͑Cr, Fe, Co, Ni, and Zn͒ doping effects in La 0.7 Ca 0.3 MnO 3 polycrystalline manganites.