Thermoelectric properties of Ca(1-x)Gd(x)MnO(3-δ) (0.00, 0.02, and 0.05) systems.

Polycrystalline samples of Ca(1-x)Gd(x)MnO(3-δ) (x = 0.00, 0.02, and 0.05) have been studied by X-ray diffraction (XRD), electrical resistivity (ρ), thermoelectric power (S), and thermal conductivity (κ). All the samples were single phase with an orthorhombic structure. The Seebeck coefficient of all the samples was negative, indicating that the predominant carriers are electrons over the entire temperature range. The iodometric titration measurements indicate that the electrical resistivity of Ca(1-x)Gd(x)MnO(3-δ) correlated well with the average valence of Mn(v+) and oxygen deficiency. Among the doped samples, Ca₀.₉₈Gd₀.₀₂MnO(3-δ) had the highest dimensionless figure of merit 0.018 at 300 K, representing an improvement of about 125% with respect to the undoped GaMnO(3-δ) sample at the same temperature.

1. Introduction

Thermoelectric generators can convert waste heat into electric energy without using moving parts and without producing carbon dioxide gas, toxic substances, or other emissions. It is expected that thermoelectric power generation can provide a new energy source from the conversion of waste heat emitted by automobiles, factories, and other similar sources. For this purpose, oxide materials are potential candidates for a wide range of high-temperature applications due to their high chemical stability and the absence of harmful elements in their compositions. Since the discovery of large thermoelectricity in NaCoO2 [1], enthusiastic efforts have been devoted to explore new oxides exhibiting high thermoelectric performances, and some layered cobaltites such as [Ca2CoO3][CoO2]1.62 and [Bi0.87SrO2]2[CoO2]1.82 are found to exhibit interesting thermoelectric properties [2-4]. These p-type materials present large thermopower (>100 μV/K at 300 K), a relatively low electrical resistivity (~1–10 mΩ–cm at 300 K), and are expected to be incorporated in thermoelectric modules. More recently, theoretical predictions lead interest to focus on low-dimensional materials with a ZT of 2.4 in Nb-doped SrTiO3 thin films due to the giant Seeback coefficient in the superlattices [5]. On the other hand, the intensive investigations on n-type oxides are still going on because their lower performances compared to p-type materials. Among these n-type materials, the perovskite CaMnO3-based compound has been intensively studied due to its relatively low electrical resistivity and high Seebeck coefficient [6-10]. The decrease in the electrical resistivity by the means of cationic substitutions in the “A” site, like in R1−AMnO3 [8] (R: rare earth cation, A: divalent cation such as Ca, Sr, Ba, and Pb), La1−SrMnO3 [9], Ca1−DyMnO2.89 [11], and Ca1−LaMnO3 [12]. Tang et al. [13] reported that Gd substitution of Ca3−GdCo4O9+ improved the figure of merit, which was about one order of magnitude larger than that of Ca3Co4O9+

In these studies, we report thermoelectric properties of Ca1−GdMnO3− (0.00, 0.02, and 0.05) systems.

2. Experimental Details

Polycrystalline samples of Ca1−GdMnO3− (0.00, 0.02, and 0.05) were synthesized by the solid-state reaction from CaCO3, Mn2O3, and Gd2O3 powders. The powders were heated at 1173 K for 10 h and at 1473 K for 20 h in air with intermediate grinding. The resulting powders were then pressed into parallelepiped and sintered in air at 1473 K for 20 h. The phase purity of resulting powders was examined by a Shimadzu XRD-6000 powder X-ray diffractometer equipped with Fe Kα radiation. Electrical resistance measurements were carried out using standard four-probe techniques. Thermopower measurements were performed between 300 and 700 K using a steady-state technique with a temperature gradient of 0.5–2 K across the sample. A type E differential thermocouple was used to measure the temperature difference between the hot and cold ends of the sample, which was measured using Keithley 2000 multimeter [14]. The temperature difference was typically between 0.5 and 1 K. The thermopower of the sample was obtained by subtracting the thermopower of Cu Seebeck probes. Thermal conductivity measurement was carried out using transient plane source techniques with very small temperature perturbations of the sample material by the hot disk thermal constants analyzer. The transient plane source technique makes use of a thin sensor element in the shape of a double spiral. The hot disk sensor acts as both a heat source for generating temperature gradient in the sample and a resistance thermometer for recording the time-dependent temperature increase [15]. The encapsulated sensor was sandwiched between two pieces of samples. During a preset time, 200 resistance recordings were taken and from them a relation between temperature and time was established. The oxygen contents and valence state of manganese were determined using iodometric titration [16].

3. Results and Discussion

XRD analysis indicates that all the samples of Ca1−GdMnO3− (0.00, 0.02, and 0.05) are a single phase with orthorhombic symmetry of Pnma. The typical XRD patterns are shown in Figure 1. The diffraction peaks are matched with earlier reports of CaMnO3 [17], and no secondary phase is observed. The similarity between the crystal structures of undoped and doped samples suggests that the doped ions do not change the crystalline structure. Lattice parameters are calculated and tabulated in Table 1. As seen in Table 1, the lattice parameters do not show monotonic trend, which may be due to the oxygen deficiency and small amount of dopants. Trukhanov et al. [18] reported that lattice parameters changed due to the oxygen deficiency in La0.50Ca0.50MnO3− (0 ≤ δ ≥ 0.50). Wiebe et al. [19] also reported that lattice parameters changed due to the oxygen deficiency in CaMnO3− (δ = 0.06,0.11).

Figure 1

XRD patterns of the Ca1−GdMnO3− (0.00, 0.02, and 0.05).

Table 1

Lattice parameters of the Ca1−GdMnO3− (0.00, 0.02, and 0.05).

Samples	a (Å)	b (Å)	c (Å)
CaMnO3−δ	5.27 (9)	7.43 (2)	5.26 (1)
Ca0.98Gd0.02MnO3−δ	5.24 (6)	7.44 (4)	5.28 (6)
Ca0.98Gd0.05MnO3−δ	5.26 (3)	7.43 (2)	5.28 (3)

Table 2 summarizes the characterization and properties of the samples at room temperature. The undoped sample shows the highest resistivity among the samples, while the resistivity of the doped samples is significantly lower than for the undoped sample due to the substitution of trivalent Gd3+ for divalent Ca2+, which decreases the concentration of holes. The value of ρ for doped samples is in the range of 0.041 Ω–cm to 0.019 Ω–cm, which increases with increasing dopant content. For doped samples, the average valence of Mn decreases and oxygen deficiency increases as compared to undoped sample. Doping of the Ca site with Gd causes a strong decrease of the ρ due to the creation of charge carrier content of Mn3+ in the Mn4+ matrix. The concentration of carriers in these samples can be correlated with the oxidation state of Mn. Creation of Mn3+ comes from two sources in the title system, that is, doping of Gd3+ and oxygen deficiency. Oxygen deficiency in CaMnO3− creates two Mn3+ five-coordinate sites for each O vacancy according to the X-ray absorption near-edge spectra results [20]. In order to compensate the oxygen deficiency and maintain the electrical neutrality, the creation of Mn3+could be present in these samples. There are also earlier reports, an oxygen deficiency for other electron-doped calcium manganites [21, 22]. The negative thermopower confirms that the dominant charge carriers are electrons for all the samples. The undoped CaMnO3− has a very large absolute S, being about −319 μV/K at 300 K. The room temperature absolute value of S for the undoped sample is lower than the value of −600 μV/K reported by Flahaut et al. [23], which may be due to the oxygen deficiency. The doped samples show relatively small absolute S due to the increase of carrier concentration (Table 2). The doped sample induces a clear decrease of the absolute S value due to the increase of the concentration of Mn3+.

Table 2

Room temperature characterization and properties of the Ca1−GdMnO3− (0.00, 0.02, and 0.05).

Samples	Mnv+	δ	ρ	S	κ total	κ el	κ ph	PF	ZT
			(mΩ-cm)	(μV/K)	(W/mK)	(W/mK)	(W/mK)	(μW/cm K2)
CaMnO3−δ	3.90 (8)	−0.04 (4)	102	−319	3.72	0.007	3.713	0.99	0.008
Ca0.98Gd0.02MnO3−δ	3.92 (8)	−0.02 (4)	041	−223	1.99	0.041	1.949	1.21	0.018
Ca0.98Gd0.05MnO3−δ	3.88 (9)	−0.03 (4)	019	−113	1.26	0.019	1.241	0.67	0.016

The temperature dependence of resistivity of Ca1−GdMnO3− (0.00, 0.02, and 0.05) is shown in Figure 2. The undoped sample exhibit nonmetallic behavior in the entire temperature range, that is, the resistivity decreases with increasing temperature (dρ/dT < 0). Similar tendency was also reported for undoped sample (CaMnO3−) [23]. The doped samples show that the resistivity increases with increasing temperature in the whole temperature range, indicating the metallic behavior (dρ/dT > 0). This is a similar behavior to that of the electron-doped manganites above the metal-insulation transition temperature [24, 25].

Figure 2

The temperature dependence of the electrical resistivity of the Ca1−GdMnO3− (0.00, 0.02, and 0.05).

Figure 3 shows the Seeback coefficient (S) as a function of temperature (300–700 K) for Ca1−GdMnO3− (0.00, 0.02, and 0.05). All samples exhibit negative values of the thermopower, which indicates that the electrons are the predominant charge carriers (n-type conduction). The absolute thermopower increases with increasing temperature and exhibit metallic behavior for undoped sample, which is contrast to Ohtaki's sample [26]. Ohtaki et al. [26] reported that absolute value of thermopower decreases with increasing temperature, typical characteristic of nonmetal-like temperature dependence. This difference should be attributed to the contribution of the oxygen deficiency [27].

Figure 3

The temperature dependence of the Seebeck coefficient S of the Ca1−GdMnO3− (0.00, 0.02, and 0.05).

For materials with more than one type of charge carrier, the diffusion thermopower can be expressed as where σ and S are the partial electrical conductivity and partial thermopower associated with the ith group of carriers, respectively. We can rewrite thermopower of CaMnO3− and as where σ in and S in are the contribution from intrinsic carriers; σ ex,defect and Sex,defect are the contribution from extrinsic carriers due to oxygen defects. Since the increase of electrical conductivity (~e −) is faster than the decrease of S (~E /k T) for semiconductors, one could expect that the second term in (2) would increase and therefore the absolute value of thermopower for CaMnO3− would increase, which should be responsible for the simultaneous increase of the electrical conductivity and absolute value of thermopower with increasing temperature.

The experimental activation energies derived from electrical resistivity and thermopower measurements are distinguishable for the title system, and the adiabatic small polaron conduction model [28] has often been invoked to account for the transport mechanism in both electrical resistivity and thermopower [24, 26, 29]. Figure 4 shows similar fitting for Ca1−GdMnO3− (0.00, 0.02 and 0.05) in the present study using the following forms ρ = ρ 0 T exp(E /k T) and S = (k /e)(W /k T+ B), where E is one-half of the energy gap between the polaronic bands, W is one-half of the polaron binding energy, B is associated with the spin and the mixing entropy, and e is the electron charge with minus sign. One would expect a decrease of absolute value of thermopower with increasing temperature and obtain a negative W when fitting thermopower data for the polaronic transport.

Figure 4

Plots of (a) In () versus 1000/T and (b) S versus 1000/T of the Ca1−GdMnO3− (0.00, 0.02, and 0.05).

The thermal conductivity is measured at room temperature, and values are presented in Table 2. Total thermal conductivity (κ total) can be expressed as where κ el and κ ph represent the electronic and lattice thermal conductivity, respectively. κ el can be calculated by using the Wiedemann-Franz-Lorenz relationship where L = π 2 k 2/3e 2 = 2.45 × 10−8 W Ω K−2 is the Lorenz number and T is the absolute temperature. κ ph is obtained by subtracting κ el from κ total. It can be clearly seen from Table 2 that the total thermal conductivity for all the doped samples is less than that of CaMnO3−. For materials with ρ > 1 Ω–cm, κ el is negligible. But in our case, the resistivity is lower than 1 Ω–cm, a fact which leads us to determine the κ el by using the Wiedemann-Franz law. The calculated value of κ el, for CaMnO3− is 0.019 W m−1 K−1 and Ca0.95Gd0.05MnO3− is 0.027 at 700 K, respectively. For all the samples, the lattice contribution is more important than the electronic one. Due to the small κ el, κ total is mainly attributed to the lattice contribution. The κ total of Ca0.95Gd0.05MnO3− is 1.26 at 300 K, being an indication of these doping effects. It should be emphasized that in contrast to the CaMnO3− and the Ca0.95Gd0.05MnO3−, samples show much lower κ total. For all doped samples, the lattice contribution is more important than the electronic one. Due to the small κ el, κ total is mainly attributed to the lattice contribution. Both of the κ el, κ ph decrease with increasing dopant content (Table 2). The radius and mass of Gd3+ and Ca2+ ions are different, and substitution of Gd3+ and Ca2+ can affect the value of κ ph. The effect of Gd3+ doping on the lattice vibration arises from two main factors. One is the crystallographic distortion and the other is the mass difference between Gd3+ and Ca2+ [24].

The power factor (S 2 σ) is calculated and presented in Table 2. The highest value of S 2 σ (1.21 μW cm−1 K−2 at 300 K) is obtained for Ca0.98Gd0.02MnO3−. The figure of merit (ZT = S 2 σT/κ) is calculated for all the samples. The calculated values are presented in Table 2. The highest ZT (0.018) is reached for Ca0.98Gd0.02MnO3−, which represents a 125% increase when compared to the CaMnO3−. Lan et al. [30] reported that the ZT value is ~0.02 at 300 K for Ca0.94Gd0.06MnO3 prepared by the coprecipitation method and ZT value is ~0.018 at 300 K for Ca0.96Gd0.04MnO3 prepared by solid state reaction, which is in agreement with our Ca0.98Gd0.02MnO3− with ZT = 0.018 at 300 K. These results suggesting that there is scope for improvement of n-type materials for high-temperature thermoelectric application.

4. Conclusions

The thermoelectric properties (ρ, S, and κ) of Ca1−GdMnO3− (x = 0.00, 0.02, and 0.05) polycrystalline samples are investigated carefully. All the samples are a single phase with an orthorhombic structure. The iodometric titration results indicate that the electrical resistivity of Ca1−GdMnO3− correlates well with the average valence of the Mn and oxygen deficiency. The smallest average valence of Mn for Ca0.95Gd0.05MnO3− has the smallest resistivity among the doped samples. The negative thermopower confirms that the domain carriers are electrons for all the samples. Doping of gadolinium on the calcium sites of CaMnO3− produces reduction of resistivity and thermal conductivity. As a result, Ca0.98Gd0.02MnO3− has the highest ZT among the doped samples. These results suggest that improving the thermoelectric properties is achived by doping concentration.