Nonmonotonic crossover in electronic phase separated manganite superlattices driven by the superlattice period

Yinyan Zhu, Biying Ye, Qiang Li, Hao Liu, Tian Miao, Lijun Wu, Lei Li, Lingfang Lin, Yi Zhu, Zhe Zhang, Qian Shi, Yulong Yang, Kai Du, Yu Bai, Yang Yu, Hangwen Guo, Wenbin Wang, Xiaoshan Xu, Xiaoshan Wu, Zhicheng Zhong, Shuai Dong, Yimei Zhu, Elbio Dagotto, Lifeng Yin, and Jian Shen

Phys. Rev. B 102, 235107 – Published 2 December 2020

Abstract

Studying manganite superlattices ${[{(LCMO)}_{2 n} / {(PCMO)}_{n}]}_{t}$ made of ${La}_{0.625} {Ca}_{0.375} Mn O_{3}$ (LCMO) and $\Pr_{0.625} {Ca}_{0.375} Mn O_{3}$ (PCMO), we found an unexpected behavior varying the period $n$ . At small $n$ , the ensemble is a three-dimensional ferromagnetic metal due to interfacial charge transfer. At large $n$ , the LCMO layers dominate transport. However, rather than a smooth interpolation between these limits a sharp transport and magnetic anomaly is found at an intermediate critical PCMO thickness $n^{*}$ . Magnetic force microscopy reveals that the phase-separation length scale also maximizes at $n^{*}$ where, unexpectedly, it becomes comparable to that of the ${({La}_{1 - y} \Pr_{y})}_{0.625} {Ca}_{0.375} Mn O_{3}$ (LPCMO) alloy. We conjecture the phenomenon originates in a disorder-related length scale: Large charge-ordered clusters as in LPCMO can only nucleate when Pr-rich regions reach a critical size related to $n^{*}$ .

Received 11 May 2020
Revised 21 August 2020
Accepted 12 November 2020
Corrected 11 December 2020

DOI:https://doi.org/10.1103/PhysRevB.102.235107

Physics Subject Headings (PhySH)

Electrical conductivity Magnetic phase transitions

Magnetic multilayers Strongly correlated systems

Magnetic force microscopy

Condensed Matter, Materials & Applied Physics

Corrections

11 December 2020

Correction: The omission of a support statement in the Acknowledgments section has been fixed. The given name of the eighth author contained an error and has been fixed.

Authors & Affiliations

Yinyan Zhu ^1,2, Biying Ye¹, Qiang Li¹, Hao Liu¹, Tian Miao¹, Lijun Wu³, Lei Li⁴, Lingfang Lin⁵, Yi Zhu¹, Zhe Zhang⁶, Qian Shi¹, Yulong Yang¹, Kai Du¹, Yu Bai¹, Yang Yu ¹, Hangwen Guo^1,2, Wenbin Wang^1,2, Xiaoshan Xu ⁷, Xiaoshan Wu⁶, Zhicheng Zhong⁴, Shuai Dong⁵, Yimei Zhu³, Elbio Dagotto^8,*, Lifeng Yin^1,2,9,†, and Jian Shen ^1,2,9,‡

¹State Key Laboratory of Surface Physics and Department of Physics, Fudan University, Shanghai 200433, China
²Institute for Nanoelectronics Devices and Quantum Computing, Fudan University, Shanghai 200433, China
³Condensed Matter Physics and Materials Science Department, Brookhaven National Laboratory, Upton, New York 11973, USA
⁴Key Laboratory of Magnetic Materials and Devices and Zhejiang Province Key Laboratory of Magnetic Materials and Application Technology, Ningbo Institute of Materials Technology and Engineering, Chinese Academy of Sciences, Ningbo 315201, China
⁵School of Physics, Southeast University, Nanjing 211189, China
⁶Department of Physics, Nanjing University, Nanjing 211189, China
⁷Department of Physics and Astronomy, University of Nebraska–Lincoln, Lincoln, Nebraska 68588, USA
⁸Department of Physics and Astronomy, University of Tennessee, Knoxville, Tennessee 37996, USA and Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, USA
⁹Collaborative Innovation Center of Advanced Microstructures, Nanjing University, Nanjing 210093, China and Shanghai Qi Zhi Institute, Shanghai 200232, China

^*edagotto@utk.edu
^†lifengyin@fudan.edu.cn
^‡shenj5494@fudan.edu.cn

Article Text (Subscription Required)

Click to Expand

Supplemental Material (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand

Issue

Vol. 102, Iss. 23 — 15 December 2020

Reuse & Permissions

Access Options

Author publication services for translation and copyediting assistance advertisement

Images

Figure 1
Intuitive explanation of expected results in our superlattices. (i) Represents bulk metallic LCMO (blue) at large $n$ , while (ii) represents just a LCMO single layer. The red curve is the idealized LCMO superlattice evolution (without PCMO) from 3D to 2D, i.e., from (i) to (ii). Naturally the resistivity increases by decreasing $n$ ; see text. Now consider our superlattice LCMO-PCMO. At small $n$ , work function differences induce charge transfer from metal to insulator, panel (iii) (yellow is PCMO metallized by LCMO). Panel (iii) becomes, thus, metallic and 3D. Its resistivity is higher than 3D LCMO, because PCMO is insulatinglike even after charge transfer. Thus, ρ(iii) > ρ(i) as observed. As $n$ grows from $n = 1$ , the PCMO region affected by charge transfer covers a smaller fraction of PCMO (red); see panel (iv). Thus, ρ(iv) > ρ(iii), as observed, defining the green curve. Combining results, the orange curve emerges, naturally containing a broad peak. The big purple arrow indicates how the effective dimension changes in the figure. While this sketch contains the essence of our results, the observed intermediate peak is much sharper than here anticipated, revealing a conjectured length scale related to quenched disorder that renders the superlattice similar to the LPCMO alloy with giant phase separation.
Reuse & Permissions
Figure 2
Transport properties of the superlattice system. (a) Temperature-dependent resistivity ( $ρ − T$ ) curves of $n$ -dependent superlattices, LPCMO and LCMO films, measured at zero magnetic field. (b) $n$ -dependent $T_{P}$ (red square) from the cooling curve and (c) low-temperature resistivity (blue up triangle) measured at 10 K for all the superlattice samples. (d) Magnetoresistance (green circle) where $MR = (R_{(0)} - R_{(H)}) / R_{(H)}$ . (e) Integral area (magenta hexagon) of $ρ − T$ curves between the cooling and warming curves, which reflects the thermal hysteresis. LCMO (orange spheres) and LPCMO (purple stars) are also marked in (b)–(e) for comparison.
Reuse & Permissions
Figure 3
Magnetic properties of the superlattice system. (a) Temperature-dependent magnetization curves (M-T). (b) Initial magnetization curves measured from 0 to 7 T after zero field cooling to 10 K. Both $M - T$ curves and initial magnetization curves are normalized for better visualization. (c) $n$ -dependent $T_{C}$ of all the superlattice samples. Inset shows the temperature-dependent magnetization (M-T) of $n = 1$ SLs measured at 100 Oe, with $T_{C}$ defined as the inflection points of $M$ ( $T$ ) (indicated by arrows as shown in inset). (d) The FM volume fraction of each sample extracted from the initial magnetization curves measured at 10 K. The inset shows the initial magnetization curves of $n = 1$ SLs measured at 10 K after cooling from room temperature under zero magnetic field. $T_{C}$ and FM volume fraction of LPCMO and LCMO films are also marked for comparison.
Reuse & Permissions
Figure 4
MFM imaging and FMM domain size comparison. (a) Mean FM domain size of each ${(LCMO)}_{2 n} / {(PCMO)}_{n}$ superlattices (red sphere) and LPCMO film (purple star). The domain size was analyzed from five images for each sample at each $T / T_{P}$ temperature. Inset shows $n$ -dependent MFM images (7 μm × 7 μm) at temperature $T / T_{P}$ under 1 T field cooling (magnetic field applied perpendicular to sample surface). Areas with negative-phase signals are the FM states (red), while areas with zero-phase signal or positive-phase signals are charge-ordered insulating (COI) states (blue). (b) Schematic side view of phase separation. (c)–(e) Top view of the corresponding simulation of EPS for (c) bulk LPCMO, (d) $n = 1$ SLs, and (e) $n = 6$ SLs. Black: FM region; yellow: AFI region.
Reuse & Permissions
Figure 5
Proposed explanation of our results. (a) At $n = 1$ , random Pr clusters are flat, and not strong enough to seed large CO clusters. (b) At large $n$ , the PCMO component (yellow) contains large Pr clusters inducing CO regions, but the LCMO component (green) dominates the metallic flow of charge. (c) At the critical $n^{*} = 6 - 7$ , the size of randomly created Pr clusters that seed large CO clusters reaches the same optimal volume (crudely, a cube of size $r^{*} = n^{*} a$ ) that also seeds the charge-ordered (CO) clusters in the LPCMO alloy (pink), as shown in (d). In both (c,d) transport is percolative. The Pr clusters, CO clusters, and electric current are marked by red cubes, transparent bubbles, and blue arrows, respectively.
Reuse & Permissions

Physical Review B

covering condensed matter and materials physics