Journal of Materials Science

, Volume 44, Issue 18, pp 4883–4891

Investigation of the quasi-ternary system LaMnO3–LaCoO3–“LaCuO3”. II: The series LaMn0.25−xCo0.75−xCu2xO3−δ and LaMn0.75−xCo0.25−xCu2xO3−δ


    • Institut für Energieforschung (IEF-1)
  • I. Arul Raj
    • Institut für Energieforschung (IEF-1)
  • Q. X. Fu
    • Institut für Energieforschung (IEF-1)
    • European Institute for Energy Research (EIfER)
  • M. Zahid
    • Institut für Energieforschung (IEF-1)
    • European Institute for Energy Research (EIfER)

DOI: 10.1007/s10853-009-3746-7

Cite this article as:
Tietz, F., Arul Raj, I., Fu, Q.X. et al. J Mater Sci (2009) 44: 4883. doi:10.1007/s10853-009-3746-7


This paper investigates the crystal structure, thermal expansion, and electrical conductivity of two series of perovskites (LaMn0.25−xCo0.75−xCu2xO3−δ and LaMn0.75−xCo0.25−xCu2xO3−δ with x = 0, 0.025, 0.05, 0.1, 0.15, 0.2, and 0.25) in the quasi-ternary system LaMnO3–LaCoO3–“LaCuO3”. The Mn/Co ratio was found to have a stronger influence on these properties than the Cu content. In comparison to the Co-rich series (LaMn0.25−xCo0.75−xCu2xO3−δ), the Mn-rich series (LaMn0.75−xCo0.25−xCu2xO3−δ) showed a much higher Cu solubility. All compositions in this series were single-phase materials after calcination at 1100 °C. The Co-rich series showed higher thermal expansion coefficients (αmax = 19.6 × 10−6 K−1) and electrical conductivity (σmax = 730 S/cm at 800 °C) than the Mn-rich series (αmax = 10.6 × 10−6 K−1, σmax = 94 S/cm at 800 °C). Irregularities in the thermal expansion curves indicated phase transitions at 150–350 °C for the Mn-rich series, while partial melting occurred at 980–1000 °C for the Co-rich series with x > 0.15.


Perovskite-type oxides, especially LaMeO3−δ with Me = Cr, Mn, Fe, Co, Ni, Cu, have been investigated by many research groups because of their interesting electrical, magnetic, and oxygen transport properties which makes them suitable for application as functional materials in solid oxide fuel cells or as giant magnetoresistant materials. In order to tailor the properties for a certain application, elemental substitution at A-sites or B-sites of the perovskite lattice is often used as this forms complex perovskite-type oxides. For instance, complex oxides in the system LaMnO3–LaCoO3 have been widely investigated in terms of their crystal structure, electrical and ionic conductivity, catalytic, magnetic and thermal expansion behavior [110]. Crystallographic studies have also been conducted on LaCoO3–LaCuO3 and LaMnO3–LaCuO3 systems [1113]. As far as we know, however, systematic studies on the complex system of LaMnO3–LaCoO3–LaCuO3 have not yet been performed.

In our previous study, the sintering behavior and properties of the oxides in the series La(Mn0.5Co0.5)1−xCuxO3−δ with x = 0, 0.05, 0.1, 0.2, 0.4, 0.6, 0.8, and 1 were investigated and the results were reported as the first part of investigations into this complex system [14]. Some remarkable results were obtained along this series of perovskite compositions: LaMn0.3Co0.3Cu0.4O3−δ crystallized as a single phase with an orthorhombic perovskite structure. Among the synthesized compositions, this compound showed the highest electrical conductivity in air at 800 °C (155 S cm−1) and the highest thermal expansion coefficient (α30–800 °C = 15.4 × 10−6 K−1). The “LaCuO3−δ” composition crystallized as a mixture of La2CuO4 and CuO at 900 °C and as a single phase with a monoclinic La2Cu2O5-type structure at 1100 °C, although previous investigations have shown that other phases are preferably formed [15, 16].

For this study, perovskites with different Mn/Co ratio were prepared and systematically characterized, i.e. seven oxides in the series LaMn0.25−xCo0.75−xCu2xO3−δ and another seven oxides in the series LaMn0.75−xCo0.25−xCu2xO3−δ with x = 0, 0.025, 0.05, 0.1, 0.15, 0.2, and 0.25 for both series. The X-ray crystallographic data, the thermal expansion coefficient data, and the electrical conductivity data were obtained on sintered samples fabricated from these oxide powders and compared as a function of the copper content.


Fourteen powders with different compositions within the series LaMn0.25−xCo0.75−xCu2xO3−δ and LaMn0.75−xCo0.25−xCu2xO3−δ (both with x = 0, 0.025, 0.05, 0.1, 0.15, 0.2, and 0.25) were synthesized by the Pechini method [17] using nitrate solutions of La, Mn, Co, and Cu in the corresponding metallic ratios. A detailed description of the synthesis process is given in Ref. [18]. After obtaining the raw powder and calcination at 600 °C for 3 h, the powders were subjected to chemical analysis using inductively coupled plasma with optical emission spectroscopy (ICP-OES, TJA-IRIS-INTREPID spectrometer) to confirm the nominal stoichiometry. In addition, two samples of each prepared powder were heat-treated in air at 900 and 1100 °C for 6 h. The crystal phase composition of these samples was determined by X-ray diffraction analysis using a Siemens D5000 diffractometer with CuKα radiation. Lattice parameters and contents of secondary phases were determined by Rietveld refinement.

The powders calcined at 600 °C were uniaxially pressed to bars (40 × 5 × 4 mm3) and sintered at either 1300 °C (for x ≤ 0.15) or 1100 °C (for x ≥ 0.2) for 6 h in air. Densities of sintered samples were measured by the Archimedean method. The thermal expansion between 30 and 800 °C was determined using a Netzsch DIL 402C dilatometer. The total electrical conductivity of the sintered samples was measured by a 4-probe DC technique at temperatures between 100 and 900 °C in air using silver wires and silver paste as contacts.

Results and discussion

Elemental analysis of the powders

Considering the precision limit (1–3%) of the analytical technique used (ICP-OES), no considerable deviations were observed between the analytical and the nominal compositions of the synthesized powders (Table 1).
Table 1

Nominal composition, analytical composition, and crystalline phases observed after calcination at 900 °C and 1100 °C for 6 h

Nominal composition

Analytical composition

Crystalline phases (numbers in parentheses indicate the phase content in wt%)

900 °C

1100 °C



Prh + La2O3 + Co3O4

Prh + La2O3 (1) + Co3O4 (1)



Prh + La2O3 + Co3O4

Prh + La2O3 (1) + Co3O4 (3)



Prh + La2O3 + L2C + Co3O4

Prh + L2C (3)



Prh + L2C + La2O3 + Co3O4

Prh + L2C (1)



Prh + L2C

Prh + L2C (3)



Prh + L2C

Prh + L2C2 (2) + L2C (7)



Prh + L2C + CuO

Prh + L2C2 (22) + L2C (2)























Por + L2C




Por + L2C



For all powders in the series LaMn0.25−xCo0.75−xCu2xO3−δ calcined at 900 or 1100 °C, in addition to the main perovskite phase, small amounts or traces of minor phases were also observed, as shown in Fig. 1. The main phase for all powders had a rhombohedral perovskite structure (Prh) similar to LaCoO3 (space group R\( \mathop 3\limits^{\_} c, \) JCPDS file no. 48-123). After calcination at 900 °C, La2O3 and Co3O4 were detected as minor phases for x = 0 and x = 0.025. With increasing Cu content (x ≥ 0.05), La2CuO4 (L2C, space group Fmmm, JCPDS file no. 38-709) appeared as a minor phase. The end member LaCu0.50Co0.50O3−δ contained a significant amount of La2CuO4 and consequently CuO was also detected. The coexistence of La2O3 and Co3O4 in compositions with x ≤ 0.1 indicates an incomplete reaction at 900 °C, which is supported by a reduced amount of these two phases after calcination at 1100 °C. For x ≥ 0.05, the La2CuO4 phase remained as the minor phase after calcination at 1100 °C with amounts of ≤7 wt% (see Table 1). For the end member LaCu0.50Co0.50O3−δ calcined at 1100 °C, the La2Cu2O5 phase (L2C2, space group C2/c, JCPDS file no. 45-10) was additionally detected with a significant amount. This result is in agreement with our previous work on the series La(Mn0.5Co0.5)1−xCuxO3−δ [14], where La2CuO4 and CuO formed at 900 °C tended to transform to La2Cu2O5 at 1100 °C.
Fig. 1

XRD patterns of LaMn0.25−xCo0.75−xCu2xO3−δ (x = 0, 0.025, 0.05, 0.1, 0.15, 0.2, and 0.25, from bottom to top) after heat treatment at a 900 °C and b 1100 °C for 6 h. The main phase with a rhombohedral perovskite structure (Prh) is indicated with open triangles. The formation of minor phases is also indicated with different symbols. Open squares: La2O3; closed squares: Co3O4; open circles: La2CuO4; closed circles: CuO; closed triangles: La2Cu2O5

The lattice parameters of the samples sintered at 1100 °C are shown in Fig. 2a. Due to the additional phases formed (Table 1), the data are not meaningful for discussing the dependence of the lattice parameters on substitution with Cu. However, in accordance with the appearance of L2C as an impurity, the lattice parameters a and c show a sudden drop between x = 0.025 and 0.05, indicating the solubility limit of Cu in LaMn0.25−xCo0.75−xCu2xO3−δ.
Fig. 2

Lattice parameters of the series LaMn0.25−xCo0.75−xCu2xO3−δ (a) and LaMn0.75−xCo0.25−xCu2xO3−δ (b) after sintering at 1100 °C

For LaMn0.75−xCo0.25−xCu2xO3−δ calcined at 900 °C, single-phase orthorhombic perovskites (Por) similar to LaMnO3 (space group Pnma, JCPDS file no. 89-2470) were obtained for x ≤ 0.15. For x ≥ 0.2, La2CuO4 was observed as minor phase. At 1100 °C, however, all powders crystallized as single-phase orthorhombic perovskites (Fig. 3). The observed crystalline phases after calcination at 900 or 1100 °C for each composition are listed in Table 1. In comparison to the Co-rich series (LaMn0.25−xCo0.75−xCu2xO3−δ), Cu apparently has a higher solubility in the Mn-rich series (LaMn0.75−xCo0.25−xCu2xO3−δ). This trend is in agreement with previous crystallographic studies on the LaCo1−aCuaO3 series and LaMn1−aCuaO3 series [1113], and can be explained by the fact that Mn ions can easily change their valence from +3 to +4 as Cu2+ ions are introduced into the perovskite lattice for the charge compensation, while only Co3+ ions exist in these compounds [13].
Fig. 3

XRD patterns of LaMn0.75−xCo0.25−xCu2xO3−δ (x = 0, 0.025, 0.05, 0.1, 0.15, 0.2, and 0.25, from bottom to top) after heat treatment at a 900 °C or b 1100 °C for 6 h. The patterns show the formation of a single-phase orthorhombic perovskite structure (Por), except in two cases (x = 0.2 and 0.25 calcined at 900 °C) which additionally contain the La2CuO4 phase as indicated by open circles

The lattice parameters of the series LaMn0.75−xCo0.25−xCu2xO3−δ do not vary significantly with increasing Cu content (Fig. 2b). Only the lattice parameter c increases very smoothly up to x = 0.15. The nearly constant lattice parameters can be understood, when the substitution with Cu is described in Kröger–Vink notation as
$$ 2 {\text{ Mn}}_{\rm{Mn}}^{ \times } = {\text{Cu}}_{\rm{Mn}}^{\prime} + {\text{Mn}}_{\rm{Mn}}^{ \bullet } $$
and the two Mn3+ ions with \( {\text{r}}_{{{\rm{Mn}}^{3 + } }} \) = 0.645 Å are replaced by a smaller and a bigger ion, i.e. Mn4+ (\( {\text{r}}_{{{\rm{Mn}}^{4 + } }} \) = 0.53 Å) and Cu2+ (\( {\text{r}}_{{{\rm{Cu}}^{2+} }} \) = 0.73 Å), respectively [19], with nearly the same mean ionic radius as \( {\text{r}}_{{{\rm{Mn}}^{3+}}} . \)

Thermal expansion

The thermal expansion behavior of LaMn0.25−xCo0.75−xCu2xO3−δ is shown in Fig. 4 and the corresponding thermal expansion coefficients (TECs) between 30 and 800 °C are presented in Fig. 5. Between 0 < x < 0.15, the TECs vary only slightly between 18.4 and 18.8 × 10−6 K−1 and a smooth maximum is observed with 19.6 × 10−6 K−1 for LaMn0.2Co0.7Cu0.1O3−δ. Only the two end members with x = 0.2 and 0.25 show a reduced TEC, which is probably due to the increased amounts of secondary phases (L2C2 and L2C). This assumption is supported by the expansion curves at high temperatures. Whereas the sample with x = 0.15 shows a small bending downwards, the two samples with x = 0.2 and 0.25 show an abrupt shrinkage starting at 980–1000 °C. Earlier studies on La0.8Sr0.2CuO2.4+δ revealed a small weight loss at this temperature due to the release of oxygen and the reduction of Cu3+ to Cu2+ [20]. The formation of a L2C phase with stoichiometric oxygen content, i.e. La2(Mn,Co,Cu)O4, might explain the small deviation from the linear expansion curve for the sample with x = 0.15. In the other two cases, however, this reaction seems to be accompanied with a partial melting of the samples. An eutectic temperature of ~1025 °C was reported in the La2Cu2O5–La2CuO4–CuO system [21], resulting in a softening of the bars and shrinkage due to the low pressure of the push-rod of the dilatometer.
Fig. 4

Thermal expansion curves for the LaMn0.25−xCo0.75−xCu2xO3−δ series
Fig. 5

Thermal expansion coefficients between 30 and 800 °C for both LaMn0.25−xCo0.75−xCu2xO3−δ and LaMn0.75−xCo0.25−xCu2xO3−δ as a function of Cu content

The thermal expansion curves of LaMn0.75−xCo0.25−xCu2xO3−δ appear rather uniform at first sight (Fig. 6a) and the TECs vary only slightly between 9.5 and 10.6 × 10−6 K−1 without any dependence on the copper content (Fig. 5). It is interesting to note that no pronounced maximum in TEC was observed for either series, whereas for La(Mn0.5Co0.5)1−xCuxO3−δ a systematic increase and decrease of TEC with a maximum at x = 0.4 was found [14]. In addition, it is worth mentioning that no shrinkage was caused by partial melting and that the Mn-rich perovskites possessed higher temperature stability than the Co-rich samples.
Fig. 6

a Thermal expansion curves for the LaMn0.75−xCo0.25−xCu2xO3−δ series. b Enlarged low-temperature region of the expansion curves

All expansion curves in Fig. 6a show a slope change at about 300 °C. An enlarged view of this low-temperature region is presented in Fig. 6b. For most of the samples, a smooth slope change was observed. For x = 0.1 and 0.15, however, a pronounced horizontal S-shape was measured. These irregularities from linear expansion are often related to phase transitions. Since all of the samples of the LaMn0.75−xCo0.25−xCu2xO3−δ series crystallized as orthorhombic perovskites, it can be assumed that they transformed to a perovskite with higher symmetry at elevated temperatures, as is the case for (La,Sr)(Co,Fe)O3 perovskites [22]. Apart from the degree of non-linearity among the different compositions, the temperature dependence of the phase transition was analyzed in more detail by generating the derivatives of the expansion curves and determining the temperatures for the onset, minimum and end of the slope change reflecting the beginning, maximum rate and end of the phase transition (Fig. 7). Depending on the Cu content, the phase transition occurred at 200 °C for LaMn0.75Co0.25O3−δ, shifted to about 350 °C for LaMn0.65Co0.15Cu0.2O3−δ and dropped back to about 150–200 °C for LaMn0.5Cu0.5O3−δ. In the derivative curves for the samples with x = 0, 0.2 and 0.25, a second minimum was found. In these cases, it is not clear whether this signal can be attributed to an additional phase change, a stepwise phase transition or the influence of inhomogeneities within the samples. It should be noted that the temperature width of the phase transition varied between 100 and 200 °C, indicating a compositional variation between different grains.
Fig. 7

Phase transition temperatures for LaMn0.75−xCo0.25−xCu2xO3−δ extracted from derivatives of the thermal expansion curves in Fig. 6b. For x = 0, 0.20 and 0.25, two minima in the derivatives of the expansion curves were observed (indicated by filled diamonds)

Electrical conductivity

The electrical conductivity of LaMnO3-δ showed a semiconducting temperature dependence and can generally be explained by the hopping of p-type small polarons [23]. On the contrary, the conductivity of LaCoO3−δ showed a semiconducting behavior at lower temperatures and a metallic behavior at higher temperatures [24, 25]. In general, the band gap between valence and conduction band is narrowing along the transitions metals from Cr to Ni [26] which explains the different temperature dependence of conductivity of the perovskites (see discussion further below). In the present work, all compositions except two showed semiconducting behavior in the temperature range studied (100–900 °C). The two exceptions, LaMn0.25−xCo0.75−xCu2xO3−δ with x = 0.20 and x = 0.25, showed metallic behavior at higher temperatures (Fig. 8a), although the sample with x = 0.25 contains 22 wt% L2C2 having semiconducting temperature dependence [14]. Therefore, we treated all samples by a hopping mechanism and considered the small impurities in the samples of LaMn0.25−xCo0.75−xCu2xO3−δ with x ≤ 0.2 (see Table 1) as negligible contribution to the total conductivity, whereas the L2C2 content in LaMn0.25−xCo0.75−xCu2xO3−δ with x = 0.25 seems to have a significant contribution to the total conductivity (see below). The temperature dependence of electrical conductivity through the hopping mechanism can be expressed according to [23] by
$$ \sigma = A/T\;{ \exp }\left( { - E_{\text{a}} /kT} \right) $$
where A is a material constant containing the carrier concentration term; Ea is the activation energy for hopping conduction; k is Boltzman’s constant; T is the absolute temperature. The plot of log σT versus 1/T for LaMn0.25−xCo0.75−xCu2xO3−δ and LaMn0.75−xCo0.25−xCu2xO3−δ is shown in Figs. 8b and 9, respectively. It can be seen that linear or nearly linear behavior was observed for both compositional series.
Fig. 8

a Measured apparent conductivity (σapp) versus temperature (T) and b log σT versus 1000/T for LaMn0.25−xCo0.75−xCu2xO3−δ. The electrical conductivity data were not corrected with the sample density. Symbols in b belong to the same samples in a. Dashed lines in b are fitted straight lines for the calculation of activation energies
Fig. 9

Log σT versus 1000/T for LaMn0.75−xCo0.25−xCu2xO3−δ. The electrical conductivity was not corrected with sample density. Dashed lines are fitted straight lines for the calculation of activation energies

From the slopes of the straight lines in the plots of log σT versus 1/T, the apparent activation energies can be calculated, as shown in Figs. 10 and 11. For LaMn0.25−xCo0.75−xCu2xO3−δ with x ≤ 0.15, the plot can be fitted with two segments of straight lines with different slopes at the lower temperature range (200–500 °C) and the higher temperature range (600–900 °C). For LaMn0.25−xCo0.75−xCu2xO3−δ with x = 0.20 and 0.25, the data points at the higher temperature range deviate too much from the linear behavior, so that only the data points at the lower temperature range were fitted. For LaMn0.75−xCo0.25−xCu2xO3−δ with x ≤ 0.10, two linear regions were found at 200–600 °C and 700–900 °C. For LaMn0.75−xCo0.25−xCu2xO3−δ with x ≥ 0.15, all data points can be fitted with a single straight line. Figure 10 shows that the activation energies for the LaMn0.25−xCo0.75−xCu2xO3−δ series at both lower and higher temperature range decrease monotonously with the Cu content. The activation energies at the lower temperature range are higher (0.32–0.22 eV from x = 0 to x = 0.15) than those at the higher temperature range (0.21–0.10 eV for the same compositional range). In contrast, the activation energies for the LaMn0.75−xCo0.25−xCu2xO3−δ series first decrease and then increase with the Cu content, as shown in Fig. 11. The minimum occurs at x = 0.10 for the lower temperature branch and at x = 0.15 for the higher temperature branch. In addition, the activation energies at the lower temperature range are lower (0.22–0.18 eV from x = 0 to x = 0.10) than those at the higher temperature range (0.35–0.22 eV for the same compositional range).
Fig. 10

Apparent and corrected electrical conductivity (σapp and σcorr) at 200, 400, 600 and 800 °C as well as activation energies for LaMn0.25−xCo0.75−xCu2xO3−δ as a function of Cu content. The relative densities of the samples are 89, 69, 96, 92, 67, 85 and 69% for x = 0, 0.025, 0.05, 0.1, 0.15, 0.2, and 0.25, respectively
Fig. 11

Apparent and corrected electrical conductivity (σapp and σcorr) at 200, 400, 600, and 800 °C as well as activation energies for LaMn0.75−xCo0.25−xCu2xO3−δ as a function of Cu content. The relative densities of the samples are 65, 62, 70, 66, 77, 85, and 90% for x = 0, 0.025, 0.05, 0.1, 0.15, 0.2, and 0.25, respectively

The dependence of the isothermal electrical conductivity on the Cu content for LaMn0.25−xCo0.75−xCu2xO3−δ and LaMn0.75−xCo0.25−xCu2xO3−δ is shown in Figs. 10 and 11, respectively. It should be noted that the sintered samples with different compositions have different relative densities ranging from 62 to 96%. In order to compare the bulk properties of samples with different compositions, the influence of the sample porosity must be taken into account. Here the as-measured apparent conductivity data were corrected with the sample density using an empirical equation according to Tagawa et al. [27]:
$$ \sigma_{\rm{app}} \, /\sigma_{\rm{corr}} = 2\left( {d_{\rm{rel}} \, / 100 \, - \, 0. 5} \right) $$
where σcorr is the corrected electrical conductivity, σapp the measured apparent conductivity, and drel the relative density (%). In Figs. 10 and 11, both the as-measured apparent conductivity and the corrected conductivity are shown as a function of Cu content. For the LaMn0.25−xCo0.75−xCu2xO3−δ series, no clear trend was found for the apparent conductivity due to interference from the sample densities. The corrected conductivity, however, increased with the Cu content at temperatures of below 600 °C. At higher temperatures, the conductivity began to decrease at x = 0.20, which can be correlated to the appearance of metallic conduction at these temperatures. In addition, the apparent conductivity of the sample with x = 0.2 was higher than the sample with x = 0.25. Besides the porosity, here the L2C2 phase may also have an influence, because this phase has an electrical conductivity of only ≈ 10 S/cm at 800 °C [14] leading to reduced conductivity values. Therefore, the compound LaCo0.5Cu0.5O3−δ should be regarded as a composite of La2Cu2O5 and Cu-substituted LaCoO3.

For the LaMn0.75−xCo0.25−xCu2xO3−δ series, both the apparent and corrected conductivity show similar profiles, i.e. a conductivity maximum is observed at x = 0.10 at the lower temperature range (<400 °C) or at x = 0.15 at the higher temperature range (>400 °C). From Figs. 10 and 11, it appears that lower activation energies correspond to higher electrical conductivity. In addition, compared to the Mn-rich series (LaMn0.75−xCo0.25−xCu2xO3−δ), the Co-rich series (LaMn0.25−xCo0.75−xCu2xO3−δ) shows significantly higher electrical conductivity, which is consistent with the fact that LaCoO3-based perovskites have a much higher electrical conductivity than LaMnO3-based perovskites [1, 10, 22]. The higher electrical conductivity of the cobaltites is due to the narrower band gap and the shifted Fermi level in the electronic band structure [26]. Although both series contain Mn and Co on the B-site of the perovskite structure, the electrical conductivity of LaMn0.75Co0.25O3−δ and LaMn0.25Co0.75O3−δ reflects the electronic band structure of LaMnO3 and LaCoO3, respectively [26]. The addition of Cu in both series leads to a higher conductivity and lower activation energy which clearly indicates the occupation of electronic states within the band gap of valence and conduction band, either as single energy levels (low Cu concentrations) or as small narrow bands (high Cu concentrations), similar to the A-site substitution with Sr in these perovskites [26]. In the case of the Co-rich series, the band gap becomes so small that the thermal energy at elevated temperatures is sufficient to activate electrons occupying energy levels in the conduction band. As shown in Fig. 8a, the maximum conductivity of the samples with x = 0.20 and x = 0.25 was observed at 500 and 400 °C corresponding to a thermal energy of RT = 6.43 and 5.56 kJ/mol (i.e. 0.067 and 0.058 eV), respectively (R = 8.3144 J K−1 mol−1). These values correspond very well with the measured activation energies at lower temperatures of these two samples (0.081 and 0.060 eV, cf. Fig. 10) and indicate a thermally activated electronic conduction in Cu-rich perovskite materials similar to Ni-based peovskites (LaNiO3, LaNi1−xFexO3) [26, 2830].


Two series of perovskite oxides (LaMn0.25−xCo0.75−xCu2xO3−δ and LaMn0.75−xCo0.25−xCu2xO3−δ with x = 0, 0.025, 0.05, 0.1, 0.15, 0.2, and 0.25) in the quasi-ternary system LaMnO3–LaCoO3–“LaCuO3” have been investigated in terms of crystal structure, thermal expansion and electrical conductivity. In comparison to the Co-rich series (LaMn0.25−xCo0.75−xCu2xO3−δ), the Mn-rich series (LaMn0.75−xCo0.25−xCu2xO3−δ) can tolerate higher concentrations of Cu in the lattice without forming Cu-containing secondary phases. After calcination at 1100 °C, all compositions in the Mn-rich series crystallized as single-phase orthorhombic perovskites showing no significant variation of the lattice parameters. The Co-rich series showed a much higher thermal expansion coefficient and electrical conductivity than the Mn-rich series. Irregularities in the thermal expansion curves indicated phase transitions at 150–350 °C for the Mn-rich series, while partial melting occurred at 980–1000 °C for the Co-rich series with higher Cu contents. For both series, the influence of the Cu content on the thermal expansion coefficient is not as pronounced as on the electrical conductivity. The electrical conductivity of the Co-rich series increased with the Cu content at temperatures below 600 °C. At higher temperatures, metallic conduction was observed for compositions with x > 0.15. For the Mn-rich series, however, a conductivity maximum was observed at x = 0.10-0.15, and all compositions showed semiconducting behavior.


The authors thank P. Lersch (deceased) and M. Ziegner for XRD measurements, A. Hilgers and M.-T. Gerhards for dilatometric measurements, and H. Lippert and N. Merki (FZJ-ZCH) for chemical analyses.

Copyright information

© Springer Science+Business Media, LLC 2009