Effect of strain and tetragonal lattice distortions in doped perovskite manganites

A series of high-quality, coherently strained $\left[{\mathrm{La}}_{2∕3}\right(\mathrm{Ca}\mathrm{or}\mathrm{Ba}{)}_{1∕3}{\mathrm{MnO}}_3∕{\mathrm{SrTiO}}_3]\times N$ superlattices has been prepared on $\left(100\right){\mathrm{SrTiO}}_3$ and ${\mathrm{NdGaO}}_3$ substrates by laser molecular beam epitaxy. The manganite layers are biaxially strained due to lattice mismatch. A quadratic decrease of the metal-to-insulator transition temperature $T_p$ with increasing biaxial distortion ${\epsilon }_{\mathrm{bi}}^2$ both for tensile and compressive in-plane strain is found. For $T>T_p$, the resistivity versus temperature curves could be well described by the small polaron hopping model with the polaron binding energy increasing with increasing ${\epsilon }_{\mathrm{bi}}^2$. Furthermore, the magnetoresistance of the manganite films was found to strongly increase with increasing ${\epsilon }_{\mathrm{bi}}^2$ or decreasing $T_p$, respectively, following a universal behavior. An anomalous upturn of resistivity in the low-temperature regime $(T<25\mathrm{K})$ was detected, which may be attributed to enhanced Coulomb interaction of the charge carriers resulting from disorder due to the lattice distortion. Our analysis clearly demonstrates the importance of biaxial strain and Jahn-Teller-type lattice distortions for the physics of the doped manganites. It is shown that epitaxial coherency strain can be used to deliberately modify the materials properties.

Figure 1

(Color online) Low-angle XRR data for a $\left[\mathrm{LCMO}\right(3.80\mathrm{nm})∕\mathrm{STO}(11.21\mathrm{nm}){]}_8$ superlattice on NGO. The upper line represents the specular XRR, the dotted line the fit to the data. The displaced, lower line is off-specular XRR data obtained for an offset angle of $0.05°$. The inset shows the rocking curve for second-order superlattice Bragg peak.

Figure 2

(Color online) Metal-to-insulator transition temperature $T_p$ vs biaxial distortion ${\epsilon }_{\mathrm{bi}}$.

Figure 3

(Color online) $\ln (\rho ∕T)$ vs $1∕T$ for different superlattices: (i) $\left[\mathrm{LCMO}\right(18.08\mathrm{nm})∕\mathrm{STO}(9.64\mathrm{nm}){]}_3$ on NGO $(E_a=0.077\mathrm{eV},{\epsilon }_{\mathrm{bi}}\simeq 0)$; (ii) $\left[\mathrm{LCMO}\right(11.96\mathrm{nm})∕\mathrm{STO}(11.42\mathrm{nm}){]}_6$ on NGO ($E_a=0.127\mathrm{eV}$, ${\epsilon }_{\mathrm{bi}}=-0.76\%$); and (iii) $\left[\mathrm{LCMO}\right(17.39\mathrm{nm})∕\mathrm{STO}(10.57\mathrm{nm}){]}_3$ on STO $(E_a=0.164\mathrm{eV},{\epsilon }_{\mathrm{bi}}=-1.82\%)$. The dependence of the polaron binding energy $E_0\approx 2E_a$ on ${\epsilon }_{\mathrm{bi}}$ is shown in the inset.

Figure 4

(Color online) Resistivity vs temperature at $H=0$ and $5\mathrm{T}$ for (a) $100\mathrm{nm}$ thick LBMO film on STO $({\epsilon }_{\mathrm{bi}}=0.17\%)$; (b) $\left[\mathrm{LBMO}\right(9.5\mathrm{nm})∕\mathrm{STO}(4.7\mathrm{nm}){]}_5$on STO $({\epsilon }_{\mathrm{bi}}=0.36\%)$; (c) the $16\mathrm{nm}$ thick LBMO film on NGO $({\epsilon }_{\mathrm{bi}}=-1.77\%)$; (d) $\left[\mathrm{LCMO}\right(17.39\mathrm{nm})∕\mathrm{STO}(10.57\mathrm{nm}){]}_3$ on STO $({\epsilon }_{\mathrm{bi}}=-1.82\%)$; (e) $\left[\mathrm{LCMO}\right(18.08\mathrm{nm})∕\mathrm{STO}(9.64\mathrm{nm}){]}_3$ on NGO $({\epsilon }_{\mathrm{bi}}\simeq 0)$; and (f) the $40\mathrm{nm}$ thick LCMO film on NGO $({\epsilon }_{\mathrm{bi}}\simeq 0)$. The solid lines are the fits to $\rho ={\rho }_0+{\rho }_2{T}^2+{\rho }_{4.5}{T}^{4.5}+a{T}^{1∕2}$.

Figure 5

(Color online) Magnetoresistance vs temperature $T_p$ for the investigated superlattices (filled triangles). Also shown are the values of bulk samples compiled in Ref. 68 (open circles). The dotted line is a fit to $\mathrm{MR}\propto (1∕T_p)\exp (E_a∕k_B{T}_p)$ with $E_a=145\mathrm{meV}$.