【作者】 师玉军；

【导师】 吴镝；

【作者基本信息】 南京大学 ， 凝聚态物理， 2013， 博士

【摘要】 钙钛矿结构的过渡族金属氧化物是一类非常吸引人的材料,源于这类材料可以展现出极为丰富的物理性质,例如从绝缘性、金属导电性到超导,从铁磁性到反铁磁性,从介电性质、压电性质到铁电性质,等等。而近些年来随着薄膜外延沉积技术特别是脉冲激光沉积(PLD)和分子束外延技术的发展,人们已经能够在原子级尺度上制备高质量的氧化物异质界面或超薄膜。在2004年,A. Ohtomo和H. Y. Hwang在两种能带绝缘体SrTiO3和LaAlO3形成的异质界面上发现了具有高迁移率的二维电子气存在,这一工作开启了在氧化物界面广泛的研究热潮。之后,更多的奇异的电子相在氧化物异质界面上相继被发现。在我攻读博士期间,我主要利用脉冲激光沉积的方法,制备研究超薄膜La1-xSrxMnO3(LSMO)的磁学性质及其对LaAl03/SrTiO3异质界面的调控。论文主要包括下面几个部分：1.金属氧化物界面及超薄膜的研究首先需要具备在原子层次上外延生长薄膜的能力。我们利用实验室自行设计的PLD系统,配以原位的反射式高能电子衍射RHEED,从生长设备的角度实现了这一首要的要求；同时,为了获得原子层平整度的界面和超薄膜,衬底也必须具有原子级平整的表面,我们可以在常用的SrTiO3衬底上经过腐蚀退火处理,得到Ti02终结的原子级平整的表面。2.对于LaAlO3/SrTiO3异质界面：在导电机制方面,通过增加覆盖层La1-xSrx MnO3,利用掺杂改变其极性的办法,我们实现了在不改变LaAlO3/SrTiO3界面迁移率的情况下,对载流子浓度的调控；通过利用生长几个单层La2/3Sr1/3MnO3抑制衬底STO在PLD生长过程中氧空缺的产生,我们构造了新的LaAlO3/SrTiO3/LSMO/STO(衬底)界面,发现界面导电不仅和LaAl03的厚度有关,还和SrTiO3的厚度有关,间接表明了LaAlO3/SrTiO3体系载流子可能来源于氧空缺。在电输运的测量中,我们观测到了明显的负磁电阻行为,进一步的分析表明该磁阻行为更可能由弱局域化导致,而非界面磁性；在利用SQUID进行的磁矩测量中,我们发现了该体系随LaAl03厚度饱和磁矩异常的变化,我们认为该信号来自氧空缺的行为,也非界面本征的磁性。3.在La2/3Sr1/3MO3超薄膜方面,我们研究了超薄膜中“死层’’(dead layer),即厚度仅几个单胞层,磁性和电导都大为退化的LSMO薄层,为此我们制备了Co/LSMO双层膜,在低温下观测到了明显的交换偏置现象,并且该交换偏置只有LSMO为1-2uc厚度时存在,表明薄膜具有反铁磁性。最后,基于死层的产生可能源于LSMO中Mn和衬底SrTiO3的Ti在界面处3d轨道发生杂化的理论,我们尝试利用生长LaAlO3间隔层的方法增强LSMO超薄膜磁性,结果与预想相反,即使是一个单胞层的LaA103也会使LSMO磁性变差。表明界面处3d轨道杂化影响LSMO薄膜磁性的理论并不成立。更多还原

【Abstract】 Perovskite transition metal oxides are a class of fascinating materials since they can exhibit a diverse range of physical properties, from insulating and metallic conduction to superconductivity, from paramagnetism and ferromagnetism to antiferromagnetism, from dielectricity and piezoelectricity to ferroelectricity. With recent advances in deposition techniques particularly pulsed-laser deposition (PLD) and molecular-beam epitaxy (MBE), people now can synthesis these materials on an atomic-scale level and make high quality of oxide heterostructures and ultrathin films. In2004, A.Ohtomo and H. Y. Hwang reported the existence of a high-mobility electron gas at the interface between two band insulators, SrTiO3(LAO) and LaAlO3(LAO). This work has intrigued a surge of researchers in oxide heterointerfaces. Many extraordinary phenomenon have been found in oxide heterointerfaces. During my Ph.D study, I have focused on LaAlO3/SrTiO3heterointerfaces and the magnetic properties of La2/3Sr1/3MnO3(LSMO) ultrathin films, which grown by using pulsed-laser deposition. My thesis mainly contains the following parts:To study such metal oxide interfaces and ultrathin films, an ability to grow epitaxial films on the atomic-scale level is primarily required. Utilizing a PLD system designed by ourselves, which combined with an in situ RHEED, we have achieved this primary requirement. On the other hand, to get well-defined interfaces or ultrathin films with atomic level flatness, the substrates must have an atomically smooth surface. By using etching and annealing methods, we have obtained atomically smooth STO(001) surfaces terminated by TiO2ionic plane.LaAlO3/SrTiO3heterointerfaces:for the mechanism of interfacial conduction, by adding La1-xSrx MnO3(LSMO-x) capping layer, whose formal polarization can be tuned by Sr doping, we can tune the carrier density of LAO/STO interface, without changing carrier mobility; utilizing an experimental fact that the formation of oxygen vacancies in STO substrates could be strongly suppressed during PLD process when STO is capped with just a few unit cells of La2/3Sr1/3MnO3film, we have constructed a new LAO/STO/LSMO/STO (substrate) interface, we found that the interfacial conduction is not only dependent on the thickness of LaA103, but also on the thickness of SrTiO3, partly indicating that carrier of LaA103/SrTiO3may be derived from oxygen vacancies in STO. For the electric transport measurements, we observed a significant negative magnetoresistance behavior. Further analysis shows that such magnetoresistance is more likely to be caused by weak localization, rather than the interfacial magnetism; For magnetic property measurements, using SQUID, we found an abnormal LaA103thickness dependence of the saturation magnetization of LaA103/SrTiO3systems, which is believed to be caused by the oxygen vacancies, not by the intrinsic interfacical magnetism.Lao.7Sro.3MO3ultrathin films:we studied "dead layer" behavior in ultrathin LSMO films. Such "dead layer" is of a few unit cells thickness and shows a large decrease of the magnetism and conductivity. We prepared Co/LSMO bilayers and observed exchange bias effect at low temperature only for the LSMO thickness is1-2unit cells, suggesting the existence of antiferromagnetic (AFM) phase in LSMO dead layer. Finally, based on the theory that the dead layer may be caused by Mn (LSMO) and Ti (STO)3d orbital hybridization at the interface, we tried to use LaA103as an interval layer to enhance the magnetism of LSMO ultrathin films. However, in contrast, LaA103interval layer causes a deterioration of LSMO magnetism, suggesting that the origin of dead layer may be much more complicated.更多还原