Chemical Bonding Center | Research Theme V: Layer-by-layer growth of multifunctional materials, and epitaxial deposition of metastable materials

NSF Chemical Bonding Center: Chemical Design of Materials

Home

Research

Theme V

Theme V: Layer-by-layer growth of multifunctional materials, and epitaxial deposition of metastable materials
Theme Leader: Salvador

The focus of this theme is the development of multifunctional materials using thin film deposition methods. The rational design of materials using thin film approaches is not traditionally an avenue of chemical research and, conversely, chemists do not often drive the choice of materials being deposited in thin films laboratories. In the CDM, thin film growth will be integrated with single-crystal and powder synthesis methods as complementary chemical approaches to realizing novel materials in forms appropriate for understanding and exploiting their physical properties. Our specific goals are

To design and grow metastable and artificially layered crystal structures that can not be synthesized as bulk materials. This has recently been realized as a technique for growing metastable, multifunctional materials, for example epitaxial stabilization allows synthesis of ferromagnetic ferroelectric BiMnO₃ films that can not be prepared at ambient pressures in the bulk[1,2]. Salvador has grown stable hexagonal YMnO₃[3] and metastable hexagonal DyMnO₃ thin films[4], and his preliminary research shows that, using appropriately engineered substrates, GdMnO₃ can be also made to adopt the metastable hexagonal form as a thin film[3]. This is particularly relevant for our work in Theme I, because compounds that both adopt the structure of YMnO₃ and contain lone pair as well as d⁰ cations are candidates for super-ferroelectricity and giant non-linear optical activity. We will use thin film synthesis to mediate the anionic order-disorder phenomena in the oxynitride phases of interest to Theme II. Substrate and synthesis conditions will be selected to favor specific structural variants (disordered, cis, trans, etc.), and the structure-property relationships will be explored. Similar thin film synthetic efforts will be carried out on materials of interest to the other themes. In all cases, density functional theory methods will be developed to predict polymorph selection during growth.

To understand how principles of chemical bonding at surfaces and interfaces generate structures with new functionalities. The chemical bond across substrate-film interfaces plays a crucial role in determining the polymorph and permits the growth of artificially layered materials[5,6,7,8]. Understanding the role of the interfacial chemical bond will be used to systematically design materials, using either stable or metastable building blocks, to generate novel polar compounds. Salvador has engineered surface structures for subsequent thin film growth[9,10,11], and has shown that details of the chemical environments in each of the constituent layers crucially control growth, the structural characteristics, and properties. In addition, residual stresses arising from substrate-film interactions can determine the degree of order or disorder in films.

In Phase I we will begin a new effort that we anticipate greatly expanding in Phase II; the layer-by-layer growth of superlattices that have important structural characteristics relevant to multifunctionality, such as the the lack of a center of symmetry (for example, in CaTiO₃/SrTiO₃/BaTiO₃ artificial structures). By depositing three different materials in sequence, such as ABCABCABC, the center of symmetry is automatically lost. By varying the orientation and chemistry of the substrate and the structural periodicity of the film, a variety of novel acentric materials that support multiple functionalities can be obtained. Layer-by-layer structures can also be used to design new ferroelectrics that have structure-property relationships arising from order/disorder phenomena. Thin film growth allows for direct control over the cationic and anionic order, along a specific direction of the unit cell, although the growth challenges can be substantial. By using chemical principles to guide materials selection, for both film and substrate structures, these challenges will be minimized for compounds likely to have exciting new properties. Although sputtering can, in principle, be used to deposit such superlattices, pulsed laser deposition or molecular beam epitaxy are more appropriate for growth of complex layered materials.

To measure properties that are experimentally inaccessible in bulk form. Some properties, such as tunability, can only be measured in thin films, since the large fields required for bulk measurements are inaccessible. In addition, to attract the interest of the wider information technology community in our materials, it is imperative that we report data (such as permittivity, magnetic permeability) for the materials in thin film form. Therefore, the ability to prepare high quality thin film samples both enhances our ability to accurately characterize the materials under study and increases the technological impact of our work.

Current research activities:

Three specific goals were articulated for this theme:

To design and grow metastable and artificially layered crystal structures that cannot be synthesized as bulk materials,
To understand how principles of chemical bonding at surfaces and interfaces generate structures with new functionalities; and
To measure properties that are experimentally inaccessible in bulk form.

Four major research activities have begun under this area:One activity involves collaboration between Salvador's and Spaldin's groups, as well as interactions with groups in Materials Science and Chemical Engineering at CMU. Graduate student Balasubramanian Kavaipatti has been investigating the stability of the multiferroic hexagonal ReMnO₃ structure as a thin film and artificially layered materials. Bala has grown thin layers and artificial structures of various ReMnO3 chemistries on various substrates using pulsed laser deposition (goal 1). Current work focuses on property measurements of hexagonal DyMnO₃, GdMnO₃, and superlattices in this family using a probe station and ferroelectric measurement system (goal 3), growth of alternate chemistries in the hexagonal structure (goal 1), and, in collaboration with Spaldin, density funtional theory calculations of relative stabilities of different crystal structures and orientations (goal Buffered substrates are being prepared to prevent strain-driven oxygen vacancies to investigate the ability to select polymorphs via epitaxial stabilization.

In a similar collaborative project, Salvador's and Woodward's groups have begun to investigate the effect of growth conditions on the stability and stoichiometry of oxynitride perovskite films. Laser ablation of A₂B₂O₇ targets (A = Sr, La, B = Ti, Zr, Sn) are being carried out in N₂ atmospheres to deposit epitaxial ABO₂N compounds on SrTiO₃ and MgO substrates (goal 1). Dielectric properties of the films will be measured (goal The long-term goal is to determine if stress in a thin film can direct anion ordering.

A third activity is a collaboration between Salvador and Stemmer to determine the influence of epitaxial film stress on oxygen vacancy ordering in oxygen-deficient perovskites. This project is building on Stemmer's expertize in sputtering stoichiometric films of (La,Sr)CoO₃; she is growing oxygen defficient (La,Sr)CoO_3-x on a range of different substrates, while Salvador is using pulsed laser deposition to deposit Sr-rich (Y,Sr)MnO₃ films, in which oxygen vacancies are believed to play a major role in stabilizing the metastable perovskite structure, on various substrates. All systems are being characterized with respect to processing conditions and substrate. Again this activity touches upon all three goals listed above. For example, SrMnO₃ is a metastable material (goal 1), chemical bonding across the interface stabilizes the perovskite structure with a specific oxygen vacancy order (goal 2), and Y-doping these metastable materials allows for magnetoresistive properties to be measured (goal 3).

Finally, Stemmer's group has focused on studies of the atomic structure of polar/non-polar epitaxial inter faces (goal 2). In particular, they have used atomic-resolution high-angle annular dark-field (HAADF) imaging in scanning transmission electron microscopy (STEM) to gain an atomic-level understanding of the structure of these interfaces. Polar/non-polar interfaces can lead to important structural and electrical phenomena at interfaces. These interfaces may contain very large charges that are not neutralized at the interface and atoms at the interface see very large fields. It is thus expected that atomic rearrangements take place at these interfaces to neutralize the interface charge. To obtain direct images of the interface, they employed HAADF imaging. In contrast to conventional high-resolution transmission electron microscopy (HRTEM) images, HAADF images are not subject to contrast reversals, and atomic column positions are obtained directly from the image. For thin samples, the image contrast is approximately proportional to the atomic number Z2. HRTEM and HAADF imaging were performed using a Tecnai F30U TEM with ultra-twin objective lens (Cs = 0.52 mm), operated at 300 kV. The following model polar/non-polar epitaxial interfaces were studied: LaAlO3/Si and ErAs/Ga(In)As. LaAlO₃ is of interest as a novel dielectric for Si technology. In contrast to the (001) Si surface, (001) surfaces of LaAlO₃ are polar. ErAs is a semi-metal with the cubic rocksalt structure. Metal nanoparticle/semiconductor composites show properties that enable new devices, such as photomixers for solid-state THz emitters. Composite properties can be engineered through controlling the density, size and arrangement of the metal particles. While bulk ErAs has the rock salt structure, it is known that ErAs nucleates in an embedded mode where Er displaces Ga and forms islands inside the zinc blende semiconductor surface. It is thus conceivable that very small ErAs particles may initially adopt the crystal structure of the zinc blende host. The (001) surface of the rock salt structure is non-polar, while that of the zinc blende semiconductor is polar, consisting of either As or (In)Ga atoms, respectively. All epitaxial heterostructures studied here were grown by molecular beam epitaxy (MBE) in collaboration with Prof. Schlom's group at Penn State (LaAlO₃/Si) and Prof. Gossard's group at UCSB (ErAs/Ga(In)As).

Reseach findings:

The stability limits of the hexagonal phase in the REMnO₃ family (RE=rare earth) have been established and extended by growing on platinized SrTiO₃ substrates.
LaAlO₃/Si interfaces exhibit an unusual 3x1 interface reconstruction, in which every third La column is removed from the interface plane. Charge compensated (001) LaAlO₃/Si interfaces, constructed using simple electron counting models, are difficult to realize experimentally. Ideally, bonding at the interface satisfies the valence requirements and produces no dangling Si bonds. Dangling bonds are half-filled (with one electron) sp₃ hybrids, and give rise to a metallic interface. Each surface atom of a (001) Si surface has two dangling bonds, one of which can be removed by a 2x1 dimer surface reconstruction. In bulk LaAlO₃, each LaO layer donates 1/2 electron to each of the two adjacent AlO₂ layers (assuming an ideal ionic crystal). To be stable, a LaO surface (unreconstructed) requires a mechanism that removes 1/2 electron per surface unit cell. Unreconstructed LaO (AlO₂) surfaces, as op- posed in the literature require the donation (ac ceptance) of 1/2 electron per surface unit cell to (from) the semiconductor and do not meet the con- ditions for charge neutrality. Given their similar charge compensation problems, the experimentally observed 3x1 reconstructed interface may be favored over an unreconstructed LaO terminated interface if Si-O bonds are energetically preferred. Al-O termi- nated interfaces may yield charge compensated in- terfaces with no dangling bonds if 3/4 of the surface oxygens were removed. Interfaces constructed from charge-neutral surfaces, i.e., with 1/6 of the La re- moved from a La-O surface or 1/4 O from an Al-O surface, may yield electrically favorable interfaces, if the Si dangling bonds were passivated with oxy- gen or hydrogen, respectively. These findings raise serious doubts whether LaAlO₃ can be used as an epitaxial gate dielectric.
The crystal structure and particle morphology of semimetallic ErAs nanoparticles embedded in epitaxial In_0:53Ga_0:47As layers has been determined. Deposition of increasing amounts of Er results in a higher density of particles and particles coalescence for high amounts. Despite overlap with the matrix through the thickness of the sample used for transmission electron microscopy, the crystal structure of ErAs particles is unambiguously identiffied as rock salt. The As sublattice is continuous across the interface between the particle and the zinc blende semiconductor. Even with a higher spatial resolution, the overlap with the matrix makes it impossible to distinguish whether the interface layer in the zinc blende semiconductor at the interface is anion or cation terminated, as both models appear identical in projection. The different possible interface terminations may, however, be distinguished in epitaxial, layered ErAs/InGaAs heterostructures. These studies are currently underway.

References

[1] A. M. dos Santos, S. Parashar, A. R. Raju, Y. S. Zhao, A. K. Cheetham, and C. N. R. Rao. Evidence for the likely occurrence of magnetoferroelectricity in the simple perovskite, BiMnO₃. Solid State Commun., 122:49-52, 2002.

[2] E. Ohshima, Y. Saya, M. Nantoh, and M. Kawai. Synthesis and magnetic property of the perovskite Bi_1-xSr_xMnO₃ thin film. Solid State Commun., 116(2):73-76, 2000.
[3] P. A. Salvador, T.-D. Doan, B. Mercey, and B. Raveau. Stabilization of YMnO₃ in a perovskite structure as a thin film. Chem. Mater., 10:2592-5, 1998.
[4] A. A. Bosak, A. A. Kamenev, I. E. Graboy, S. V. Antonov, O. Yu Gorbenko, A. R. Kaul, C. Dubourdieu, J. P. Senateur, V. L. Svechnikov, H. W. Zandbergen, and B. Hollander. Epitaxial stabilization of hexagonal RMnO₃ (R = Eu-Dy) manganites. J. Mater. Chem., 12:800-1, 2002.
[5] P. A. Salvador, T.-D. Doan, Mercey B., and B. Raveau. Thin film synthesis of copper-based perovskites having two-dimensional cation order: (La_0.8Ba_0.2CuO_2.6Â±6)_m(AECuO₂) superlattices. pages 285-290. Materials research society, 1999.
[6] B. Mercey, P. A. Salvador, W. Prellier, T.-D. Doan, J. Wolfman, J. F. Hamet, M. Hervieu, and B. Raveau. Thin film deposition: a novel synthetic route to new materials. J. Mater. Chem., 9:233-244, 1999.
[7] B. Mercey, A. M. Haghiri-Gosnet, M. Hervieu, Ch. Simon, D. Chippaux, Ph. LeCoeur, W. Prellier, P. A. Salvador, and B. Raveau. In-situ monitoring of the growth and characterization of (PrMnO₃)_n(SrMnO₃)_n superlattices. J. Appl. Phys., 94:2716-2724, 2003.
[8] P. A. Salvador, A.-M. Haghiri-Gosnet, B. Mercey, M. Hervieu, and B. Raveau. Growth and magnetoresistive properties of (LaMnO₃)_m(SrMnO₃)_n superlattices. Appl. Phys. Lett., 75:2638-2640, 1999.
[9] P. A. Salvador, B. Mercey, O. Perez, A.-M. Haghiri-Gosnet, T.-D. Doan, and B. Raveau. Growth and structural characterization of Sr₂TiO₄: Chemical control over the terminating SrTiO₃ surface. Mater. Res. Soc. Symp. Proc., 587:1-6, 2000.
[10] A. Asthagiri, C. Niederberger, A. J. Francis, L. M. Porter, P. A. Salvador, and D. S. Sholl. SrTiO₃(111) surface: An experimental and theoretical study. Surf. Sci., 537:134-152, 2003.
[11] T.-D. Doan, J. Giocondi, G. S. Rohrer, and P. A. Salvador. Surface engineering along the close packed direction of perovskite oxides. J. Cryst. Growth, 225:178-182, 2001.