Project supported by the National Natural Science Foundation of China (Grant No. 61471301).
Project supported by the National Natural Science Foundation of China (Grant No. 61471301).
† Corresponding author. E-mail:
Project supported by the National Natural Science Foundation of China (Grant No. 61471301).
The magneto-electric effect in magnetic materials has been widely investigated, but obtaining an enhanced magneto-electric effect is challenging. In this study, tricolor superlattices composed of manganese oxides—Pr0.9Ca0.1MnO3, La0.9Sr0.1MnO3, and La0.9Sb0.1MnO3—on (001)-oriented Nb:SrTiO3 substrates with broken space-inversion and time-reversal symmetries are designed. Regarding the electric polarization in the hysteresis loops of the superlattices at different external magnetic fields, both coercive electric field Ec and remnant polarization intensity Pr clearly show strong magnetic-field dependences. At low temperatures (< 120 K), a considerable magneto-electric effect in the well-defined tricolor superlattice is observed that is absent in the single compounds. Both maxima of the magneto-electric coupling coefficients ΔEc and ΔPr appear at 30 K. The magnetic dependence of the dielectric constant further supports the magneto-electric effect. Moreover, a dependence of the magneto-electric effect on the periodicity of the superlattices with various structures is observed, which indicates the importance of interfaces. Our experimental results verify previous theoretical results regarding magneto-electric interactions, thereby paving the way for the design and development of novel magneto-electric devices based on manganite ferromagnets.
Since the magneto-electric (ME) effect—induction of magnetization by an external electric field or induction of electric polarization by an external magnetic field—was theoretically confirmed in 1959–1960 by Astrov,[1–3] it has been the hottest topic in the field of condensed matter physics for a long period. During the last half of the 20th century, ME effects have been observed in many single-phase materials and composites.[4–7] For example, Wang et al. reported the electric and magnetic properties and the ME effect in Ba0.8Sr0.2TiO3/CoFe2O4 heterostructure films.[8] Unfortunately, the magnitude of the observed ME effect is too low to meet the application requirements of practical devices.
Two or all three ferroic orders (ferroelectrics, ferromagnetics, and ferroelastics) can coexist in multiferroic materials.[9–12] Currently, the research of new materials with gigantic ME coupling is attracting a great deal of attention. Two main reasons exist for the focus on multiferroic materials. First, multiferroics with simultaneously coupled magnetic and electronic degrees of freedom offer an additional degree of freedom for the design of many advanced functional devices, e.g., sensors, drivers, or storage devices.[13–15] Second, owing to the coupling between ferromagnetic and ferroelectric states, certain novel features absent in the single states can be expected.[16] Unfortunately, such materials are rare in nature because the ferroelectricity and magnetism need different broken inversion symmetries in micro-electronic states. Although new multiferroics have been reported, their ME coupling is limited or the properties cannot appear at room temperatures.[17] Therefore, obtaining strongly coupled electrical polarization and magnetization at room temperatures is still a great challenge, and more research on multiferroic materials should be conducted to improve the ME effect. Owing to the rapid development of the thin-film deposition technology, it is possible to grow atomic-scaled flat films. This development encourages theoretical investigations and leads to the emergence of numerous artificial materials.[18–20] Controlling the growth progress of dissimilar materials at an atomic scale is expected to obtain fascinating enhanced and novel properties. At present, many different approaches are used to increase the ME coupling, e.g., chemical doping, highly oriented epitaxial films, or heterojunction composite films composed of ferroelectric and ferromagnetic layers.[21–24] Among these approaches, superlattices (SLs) can be viable candidates. They are composed of thin layers of two or more different compositions that are stacked in a specific sequence. This system can produce striking properties that do not exist in its constituent compounds. Norton et al. (1994) reported the superconductivity in SLs composed of non-superconducting SrCuO2 and BaCuO2.[25] Kida et al. (2007) reported an enhanced optical ME effect in a patterned artificial tricolor SL.[26] Furthermore, Tokura predicted that non-ferroelectrics can give rise to a ferroelectric state in a magnetically ordered state in a fabricated tricolor SL with artificially broken symmetry.[27] This hypothesis was confirmed in our early studies on frustration-induced ferroelectricity in manganite tricolor SLs.[28] Moreover, perovskite manganese oxide exhibits a strong correlation and coupling system regarding the degrees of freedom of orbit, spin, and lattice, which can lead to the emergence of certain electric and magnetic states and physical properties.[29,30]
In our previous study, we have investigated a tricolor SL composed of non-ferroelectric La0.9Sr0.1MnO3 (LSMO, A), Pa0.9Ca0.1MnO3 (PCMO, B), and La0.9Sb0.1MnO3 (LSbMO, C), which exhibits a multiferroic state below 30 K.23 The ferroelectricity of SLs on periodicity has shown that interfaces play a key role for ferroelectricity and that the ferroelectric properties of SLs vary with the number of interfaces.
In the present study, we concentrate on the role of a magnetic field on the ferroelectric polarization of tricolor SLs. Therefore, a series of SLs [(LSMO)n/(PCMO)n/(LSbMO)n]m were fabricated by conducting laser molecular-beam epitaxy (L-MBE) deposition, where n is the thickness of each sublayer and m is the periodicity of the alternating cycle. The SL structure is ABCABC-stacked and labeled by (n, m).
The SL film growth was carried out at 1053 K with 0.5 Pa of flowing oxygen targeting the (001)-oriented Nb:SrTiO3 (NSTO) single-crystal substrate. The total sample thickness was fixed to 51.25 nm and controlled with a transmission electron microscope (TEM) (JEM-2100 F, JEOL). The electric polarization P versus the electric field E (P–E), i.e., the hysteresis loops of the SL films, are measured with a modified Sawyer–Tower circuit (Precision LC, Radiant Technologies) and by placing the samples into a closed-circuit cryostat (C300, Janis Research Company). The dielectric properties of the SL films are calculated from the capacitance values measured with an LCR meter (E4980 A, Keysight Technologies). More details on the sample preparation and structural characterization of SLs can be found in Ref. [28].
To measure the electrical properties of the SL films, aurum electrodes are fabricated on their surfaces using a shadow mask to obtain ohmic contacts. The size of each aurum electrode is 1 × 1 mm2. The measuring current is perpendicular to the in-plane direction. Figure
Figure
In a multiferroic, the dielectric constant, like the polarization, is coupled to the magnetic order. Since the magnetic field affects the magnetic order, the field also indirectly alters the dielectric constant of ME multiferroics.[34,35] Therefore, the magneto-capacitance effect can reflect the ME effect to a certain extent. As shown in Fig.
To gain more knowledge on the observed MD effect, the dielectric constant and loss versus temperature are measured at 2 kHz under 0 kOe, 5 kOe, and 10 kOe, as shown in Fig.
In order to understand the true nature of the ME effect, P–E hysteresis loops of a series of tricolor SLs with various (n, m) are measured. Figure
We successfully fabricated [(LSMO)n/(PCMO)n/(LSbMO)n]m tricolor SL films on (001)-oriented Nb:SrTiO3 single-crystal substrates with (L-MBE) at 1053 K to investigate the ME and MD effects of SL films. According to the electric-polarization hysteresis loops at different external magnetic fields, both coercive electric field and remnant polarization intensity of the SLs show strong dependences on the magnetic field, suggesting a strong positive ME effect that does not appear in the single-phase compounds. Further, a negative MD effect is observed at low temperatures, which indirectly supports the ME effect. Both ME and MD effects depend on (n, m). Consequently, the interfaces contribute to an increased ME effect. The maximum ME coupling coefficient in the SL is obtained with (n, m) = (3, 15) at 30 K. Our results confirm that the ME effect can be increased by employing asymmetric tricolor SLs—even though the SL components are non-multiferroic materials. These results pave the way for the design of novel ME devices.
[1] | |
[2] | |
[3] | |
[4] | |
[5] | |
[6] | |
[7] | |
[8] | |
[9] | |
[10] | |
[11] | |
[12] | |
[13] | |
[14] | |
[15] | |
[16] | |
[17] | |
[18] | |
[19] | |
[20] | |
[21] | |
[22] | |
[23] | |
[24] | |
[25] | |
[26] | |
[27] | |
[28] | |
[29] | |
[30] | |
[31] | |
[32] | |
[33] | |
[34] | |
[35] | |
[36] | |
[37] |