Plasma augmented structural and electrical properties of half doped neodymium strontium manganites

Powder samples of polycrystalline Nd_0.5Sr_0.5MnO₃ were synthesized using a conventional solid-state reaction method. High purity (>99.99%) powder of Nd₂O₃, MnO₂, SrCO₃ were preheated at around 800 °C for 5 h. These powder were then taken in the stoichiometric proportion and subjected to rigorous grinding for 5–6 h. The grounded mixture was then subsequently calcined at 1100 °C for 24 h. The calcination was repeated thrice with intermittent grindings. After the final calcination, the calcined product was re-grounded and pressurized to form pellets. Sintering of the pellets was done at 1250 °C for around 36 h.

The pellets were then exposed to oxygen plasma for a duration of 60 s. The conventional glow discharge method was used to produce oxygen plasma. The plasma chamber had a height of 50 cm and a diameter of 40 cm wherein the working pressure was maintained at 1.7 × 10⁻¹ mbar by introducing the oxygen gas and the discharge condition was achieved using a proper power source. The typical plasma density and temperature as measured by plane Langmuir probe were 10⁸–10⁹ cm⁻³ and 1–3 eV respectively. Then the samples were kept on a target holder and placed inside the plasma system, where there is almost uniform plasma density. The sample was treated with oxygen plasma at room temperature for 60s under a uniform discharge condition.

Characterization is an essential technique to evaluate the crystallinity, phase, and structural parameters of the samples. XRD data of unexposed and plasma exposed samples were recorded in the range of 20°–80° with a step size of about 0.02° using Rigaku Miniflex diffractometer with the x-ray source as Cu Kα radiation (λ = 1.54 Å). EDS study augmented our result by confirming the stoichiometric proportion of the sample. EDS measurement was carried out with EVO MA18 with Oxford EDS(X-act) at room temperature. For x-ray photoelectron spectroscopy (XPS) measurement PHI 5000 VERSAPROBE II was employed, which validated the surface composition of the samples and determined the oxidation states of the Mn species in the unexposed and plasma exposed sample under investigation.

The four probe method was employed for the measurement of the dc electrical resistivity within the temperature range 10–300 K in a closed cycle refrigerator (CCR). A constant current of 1 mA was provided through the current terminals using Keithley Current Source (Model 6221), while the voltage across the voltage terminals was measured by Keithley Nanovoltmeter (Model 2182A). In addition, another significant transport property i.e. the Seebeck coefficient (S) was recorded in the temperature range 5–300 K, using differential DC method.

Rietveld refinement using FullProf program was employed to analyze the XRD data. The fitting is shown in figure 1. From the analysis, the lattice parameters of the unexposed and plasma exposed samples were derived. The analysis also validates the purity and phase of the sample within experimental measurements. Both unexposed and plasma exposed NSMO samples have orthorhombic structure and Pbnm symmetry. Hence, we can conclude that the symmetry of the system is unaltered by plasma exposure. Table 1 enlists the values of structural parameters of the samples as derived from XRD analysis. The lattice parameter, a decreases while b and c increases as a result of plasma exposure.

Zoom In Zoom Out Reset image size

From table 1 it is evident that there is a reduction in the particle size post plasma exposure. This in turn increases the surface area, which leads to an increase in cell volume [30]. Plasma induces a surface modification in the range of 20∼200 nm depending upon the energy of plasma as observed in most of the cases. The oxygen plasma used comprises of charged species along with oxygen radicals and energetic photons, which is responsible for a reduction in the particle size. Plasma is an established technique used in the synthesis of nanoparticles which can easily reduce the particle size depending upon the time of exposure [31].

Bond angle and bond length were also generated from the FullProf program, which has been tabulated in table 1, which shows that the apical bond length Mn–O1 as well as the basal bond length Mn–O2 diminishes due to plasma treatment. As a result, the tilt angle {180-(Mn–O–Mn)}/2 between two adjacent MnO₆ octahedra decreases.

Hence, the Jahn-Teller distortion of the octahedra weakens [32]. On the contrary, enhancement of the apical bond angle Mn–O1–Mn and basal bond angle Mn–O2–Mn is observed due to plasma treatment. The crystallite size has been measured for both the pristine and plasma exposed (PE) samples using the well-known Debye–Scherrer (DS) formula and Williamson-Hall (WH) plot method. Debye–Scherrer formula is written as

$$\begin{eqnarray}&&D=\displaystyle \frac{K\lambda }{\beta \,\cos \,\theta },\end{eqnarray} \tag{ 1 }$$

where D is crystallite size, K (=0.94) is the shape factor, λ (=1.54 Å) is the wavelength of x-ray radiation, and $\beta$ is the full-width half maxima of the most intense diffraction peak corresponding to the Bragg angle 2θ. DS method accounts for the XRD peak broadening only due to crystallite size. So, we also adopt the WH plot method to determine the crystallite size. In this method, the XRD peak broadening is attributed to the crystallite size as well as the lattice strain. WH formula is represented as

$$\begin{eqnarray}&&\beta \,\cos \,\theta =\displaystyle \frac{k\lambda }{D}+4\varepsilon \,\sin \,\theta ,\end{eqnarray} \tag{ 2 }$$

where k = 0.75 (for spherical particles) and $\varepsilon$ is the strain [33]. Figure 2 shows the WH plot for both pristine and plasma exposed samples. It is observed from both methods (DS &WH) that plasma exposure leads to a decrease in the crystallite size. The strain is computed from the slope of the WH plot, and it is observed to increase as a result of plasma exposure.

Zoom In Zoom Out Reset image size

EDX measurement was carried out to ascertain the compositional purity of the pristine sample. The measured compositions of the elements were computed as given in table 2, which is found to be close to its nominal value. The analysis supports the accomplishment of the synthesis of the sample with no significant stoichiometric loss of elements within experimental limits.

The wide survey scan of both the unexposed and plasma exposed NSMO samples, depicted in figure 3, validate the chemical composition of NSMO sample with the peaks of Nd3d, Sr3d, Mn2p, O1s, and the adventitious C1s peaks at binding energy values, which are in close agreement with previous reports [34]. The C1s peak appears owing to the unavoidable contaminant carbon on the specimen's surface. A detailed study has been done on the core energy level XPS data for Mn2p, Nd3d, Sr3d, and O1s. The doublet Mn2p spectrum as shown in figure 4 for the UE sample is attributed to the spin-orbit splitting into Mn2p_3/2 and Mn2p_1/2 peaks approximately at 641.8 eV and 653.25 eV, respectively. The separation between these peaks is about 11.45 eV, which is in good agreement with earlier reports [35]. Analysis of the doublet Mn2p spectrum confirms the occurrence of mixed valence states (Mn³⁺ and Mn⁴⁺) of Mn ions. Mn³⁺ and Mn⁴⁺ peaks are observed to be located at 641.4 eV and 642.8 eV respectively. The atomic concentration of Mn³⁺ and Mn⁴⁺ is found to be 51.4% and 48.6% thereby rendering the ratio Mn⁴⁺/Mn³⁺ as 0.945, which is very close to the nominal value of 1 in the case of half-doped manganites [4, 36]. Plasma exposure does not seem to affect the peak position of Mn⁴⁺ and Mn³⁺ peaks significantly. However, Mn³⁺ and Mn⁴⁺ concentrations are found to be 44.1% and 55.9% respectively making the Mn⁴⁺/Mn³⁺ ratio as 1.26 which hints toward an oxygen rich sample [37, 38].

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Three peaks are required to properly fit the O1s core spectrum, as seen in figure 5(a), for the UE NSMO sample at 529.1 eV, 530.6 eV, and 531.8 eV corresponding to Mn–O, Nd–O, and Sr–O bonds respectively based on Barr's classification of oxides [39]. Similar fitting of the O1s core spectrum shown in figure 5(b) for the PE NSMO sample depicts that plasma exposure does not vary the peak positions much as Nd–O, Mn–O, and Sr–O bonds are fitted at 530.22 eV, 528.6 eV, and 531.66 eV respectively. However, it is observed from the fitting, that Mn–O content is largest for the unexposed sample while Nd–O and Sr–O content increases for the plasma exposed samples. The increase in the Mn⁴⁺/Mn³⁺ ratio, as well as the increase in Nd–O and Sr–O content, as a consequence of plasma exposure, is a clear indication of oxygen non-stoichiometry, which is mainly due to the incorporation of oxygen due to plasma exposure. Curve fitting of the core spectra of Sr shown in figures 6(a) and (b) predicts the existence of SrO and SrCO₃ while that of Nd given in figures 6(c) and (d) predicts the existence of Nd₂O₃, which may be from the lattice of the sample itself. In the case of Nd 3d spectrum, plasma exposure does not affect the peak position of Nd₂O₃ much but the O KLL peak corresponding to an Auger transition becomes prominent [34]. Post plasma exposure, the peak positions of all components of Sr 3d spectrum are not affected much, while the peaks become sharper.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Temperature dependent electrical resistivity ρ(T) for the untreated and plasma treated samples of NSMO is shown in figure 7. Insulating behavior is depicted for the untreated sample, which is consistent with other reports [22, 40]. Through the entire range of temperatures, there is no indication of the metal-insulator transition. From the inset of figure 7, it is interesting to note that post plasma exposure, the behavior remains unchanged, though there is a drastic reduction of resistivity at all temperatures for the plasma treated sample. This sharp reduction in the resistivity may plausibly be attributed to the elongation of the Mn–O–Mn bond angle (as depicted in table 1) due to plasma treatment.

Zoom In Zoom Out Reset image size

Due to the stretching of the bonds, the prospect of charge carrier hopping between the Mn³⁺ and Mn⁴⁺ ions enhances, thereby boosting the conductivity. Further, as a result of oxygen plasma treatment, oxygen perhaps diffuses into the sample, which also contributes to the reduction in resistivity, which is corroborated by the XPS analysis. In order to reveal the conduction mechanism involved in our investigated samples, we have considered the two well-known models (i) small polaron hopping (SPH) model [41] for high temperatures and (ii) variable range polaron hopping (VRH) model [42] for low temperatures. The isolation of the two temperature zones has been done in terms of Debye temperature θ_D, which has been derived from the deviation of the straight line fit of the SPH model [43] as seen in figure 8. SPH model is valid in the temperature zone $T\geqslant \tfrac{{\theta }_{D}}{2}$ and mathematically represented as

$$\begin{eqnarray}&&\rho ={\rho }_{0}T\exp \left(\displaystyle \frac{{E}_{A}}{{k}_{B}T}\right),\end{eqnarray} \tag{ 3 }$$

where E_A signifies the activation energy, T the absolute temperature, and ρ₀ the residual resistivity. Figure 8 exhibits that the variation of ln (ρ/T) w.r.t 1/T, which is a straight line thereby confirming that small polarons are involved in the conduction process for the sample when $T\geqslant {\theta }_{D}/2.$ We extract the value of the activation energy from the slope of the straight line fit. It is seen that the activation energy increases with plasma treatment. For $T\lt {\theta }_{D}/2$ we use the VRH model which is expressed as

$$\begin{eqnarray}&&\rho ={\rho }_{0}\exp {\left(\displaystyle \frac{{T}_{0}}{T}\right)}^{1/4}.\end{eqnarray} \tag{ 4 }$$

Zoom In Zoom Out Reset image size

Figure 9 shows that this model is applicable in the temperature zone $T\lt {\theta }_{D}/2.$ The slope of this curve gives the value of T₀ and the density of states at the Fermi level N(E_F ). These two parameters are related through the equation

$$\begin{eqnarray}&&N\left({E}_{F}\right)=\displaystyle \frac{18{\alpha }^{3}}{{k}_{B}{T}_{0}}.\end{eqnarray} \tag{ 5 }$$

Here, α is the electron wave function decay constant, which quantifies the confinement of charge and k_B is the Boltzmann constant [44]. N(E_F) is observed to increase as a result of plasma treatment.

Zoom In Zoom Out Reset image size

The behavior of thermoelectric power (TEP) S for the pristine and PE NSMO samples is shown in figure 10. In the entire temperature range of investigations S, for the pristine as well as plasma treated sample, is negative implying the existence of electrons as the predominant charge carriers. It is observed that the absolute value of S increases with an increase in temperature in the case of an unexposed sample. Further, the plasma exposed sample also exhibits a similar variation of S with temperature. However, plasma treatment decreases the absolute value of TEP for high temperature (>225 K) while it is increased for low temperature.

Zoom In Zoom Out Reset image size

Analysis of the thermoelectric data complements the information regarding the conduction mechanism provided by the resistivity data. Therefore, similar to resistivity, thermoelectric data is analyzed using SPH and VRH models in two different temperature regimes. SPH analysis (figure 11) is done at the high temperatures, which is governed by the following equation

$$\begin{eqnarray}&&S=\displaystyle \frac{{k}_{B}}{e}\left(\displaystyle \frac{{\rm{\Delta }}}{{k}_{B}T}+\alpha \right),\end{eqnarray} \tag{ 6 }$$

where k_B is the Boltzmann constant, Δ is the activation energy and α is a constant, which is related to the entropy of the charge carriers [41]. The computed values of Δ and α is given in table 3. Small polarons are responsible for the transport properties in both the pristine and plasma treated samples as α < 1.

Zoom In Zoom Out Reset image size

Table 3. Fitting parameters obtained from resistivity and thermoelectric power data for NSMO samples.

	SPH Model fitting for resistivity		VRH Model fitting for resistivity			SPH Model fitting for thermopower		VRH Model fitting for thermopower
Dosage	EA (meV)	${\rho }_{\infty }\times {10}^{-5}$ (Ω-mm)	T0 × ${10}^{4}$ (K)	${\rho }_{0}$ (Ω-mm)	N(EF) × ${10}^{28}$ (eV−1 m−3)	Δ (meV)	α	S0 (μV/K)	K (μV/K1/2)
Pristine	76.5	53.4	3.54	0.44	6.45	12.49	−0.98	5.18	1.73
Plasma exposed	95.4	0.60	1.27	0.93	1.70	8.46	−0.77	2.67	1.73

Further, it can be seen from Table 3 that the activation energies obtained by analysing the resistivity data (E_A ) are much larger than that from thermopower data (Δ). This large difference suggests SPH conduction in the investigated samples [45]. The difference in the results is because part of the activation energy E_A is designated for the creation of the carriers and the rest (Δ) is utilized to set off it's hopping [45]. The activation energy, Δ is observed to diminish due to plasma treatment, which boosts the conductivity in the PE sample. For low temperatures akin to the resistivity analysis we use the VRH model, which defines the thermoelectric power as

$$\begin{eqnarray}&&S={S}_{0}+K{T}^{1/2},\end{eqnarray} \tag{ 7 }$$

$$\begin{eqnarray}&&K=\left({k}_{B}/3e\right){\xi }^{2}{T}_{0}^{1/2}\left[\displaystyle \frac{d\,\mathrm{ln}\,N\left({E}_{F}\right)}{d{E}_{F}}\right],\end{eqnarray} \tag{ 8 }$$

where ξ is a constant that determines the energy width of the conducting states [46]. Figure 12 conforms to the VRH model of thermopower at low temperatures and the fitting parameters are depicted in table 3.

Zoom In Zoom Out Reset image size