Breakthroughs in the synthesis, epitaxial growth, and characterization of complex heterostructures have in recent years brought material science to an entirely new level, allowing complex artificial structures to be realized with atomic-level precision. High-quality ultrathin films of ferroelectrics, high-temperature superconductors and magnets can be interfaced or even grown directly on silicon, thus offering tremendous new possibilities for artificial multifunctional materials and devices.
The suited combination of chemically different but epitaxially compatible materials configures multilayers and heterostructures with specifically designed properties.
In this context, the role of theoretical prediction is destined to become more and more instrumental in heading the experimentalists towards the most promising among the endless number of combinations.
Novel functionalities recently observed in surfaces, thin films, heterostructures, and superlattices promise to burst the implementation of radically new designs in microelectronics, spintronics, photovoltaic and thermoelectric energy production technology. Paramount examples are the 2D electron gas confinement at the interface of SrTiO3 and LaAlO3 perovskites, which ignited the perspective of a novel oxide-based nanotechnology, and the methilammonium lead-iodide perovskite CH3NH3PbI3, prototype of a new class of low-cost photoconversion devices with exceptional efficiency, which promise to burst a new age of solar cell technology with unprecedented perspectives.
In synergy with a worldwide network of experimental and theoretical collaborators, our future activity will point to:
– develop innovative methods, based on multi-scale integration of First-Principles plus model theories, accurate for solids and molecules alike, capable to describe the widest range of properties: structural, electronic, magnetic, optical, transport, thermodynamic, disorder.
– use them to design a new generation of materials and heterostructures with built-in functionalities, such as thermoelectric, photoconversion, charge confinement, magnetoelectric, transport and spin-transport capability.
Highly-efficient conversion of temperature gradients into electric power (Seebeck effect) and magnetization (spin-Seebeck) may allow unprecedented capabilities in term of recycling of dissipation energy and consequential reduction of power consumption, up to visionary applications such as self-powered memory elements. There is mounting evidence, that nanostructured systems (e.g. oxide heterostructures) can overcome bulk systems and provide a new generation of highly efficient thermoelectric devices: on the one hand, carriers confinement in nanostructures may favour an increase of thermopower, provided that specific band structure characteristics are matched; on the other hand, nanostructurization can reduce thermal conductivity, which plays against the net thermoelectric efficiency.
Activity of AF and his group in the field started in the last few years, addressed to deliver a rigorous description and clear understanding of the mechanism at the basis of the thermoelectric behaviour in oxides, and general guidelines for the design of heterostructures with improved thermoelectric properties. Important results for thermopower in various 2DEG applications,, were obtained in the framework of the EU-FP7 “OxIDes” Project, placing AF and his group at the forefront of the search for novel 2D systems with enhanced thermoelectric capabilities.
A fast-rising frontier in photovoltaic applications is the use of inorganic and hybrid organic/inorganic materials as carrier generators in solar cells. Innovative approaches now allow these systems to be grown by relatively inexpensive solution-based (e.g. printing) techniques at low-temperature, and be potentially suited for competitive mass-production solar cell technology. A number of characteristics must be satisfied for these materials to be suited for photoconversion applications: appropriate band gaps, large optical absorption in visible/near IR solar spectrum, transport capabilities, structural matching with the scaffold, appropriate band offsets if assembled in heterostructures. To the aim, a theoretical activity which can reliably predict structural, electronic, optical and transport properties of heterostructures can play an invaluable role in guiding experiment towards the most promising composite materials for solar cells applications. The advanced methodologies developed by AF and co-workers (specifically the VPSIC theory,) has the potential to exert a great impact in the field, overcoming the typical shortcomings of standard ab-initio theories, and accurately describing the fundamental properties of generic heterostructures.
Although started very recently, the AF activity in the field is soon becoming a primary interest for AF, given the outstanding technological implications. The focus is currently on methylammonium lead-triiodide perovskites which are inflaming the community of solar cell applications due to the recently reported record-high 15% efficiency, and the striking capability of functioning as both carrier generator and carrier conductor. Understanding these magic functionalities will help to drawing design rules for alternative systems with improved performances to be suggested to the experimental activity. The activity in the field will be carried out in synergy with an international network of partners, established in the framework of recently submitted projects at H2020-EU level and Regional level, currently under evaluation. Materials of interest for future investigations will also include copper-indium-gallium-selenide (CIGS) and copper-zinc-tin-sulphide (CZTS) semiconductors, pnictogen sulfide Bi2S3, Sb2S3 nanoribbons, ZnO-polymer thin films.
Since its first observation by Ohtomo and Hwang in 2005, an exceptional flurry of scientific activity has been addressed to the study of 2DEG at the SrTiO3/LaAlO3 (STO/LAO) interface. This unexpected discovery opened the route towards exceptionally promising applications in electronics, optoelectronics and energy conversion devices. The system was studied by AF in the framework of the FP7 EU “OxIDes” Project, a collaboration of a large network of experimental and theoretical collaborators (Jean-Marc Triscone in Genève, Jochen Mannhart in Augsburg, Philippe Ghosez in Liège, Javier Junquera in Santander). The project delivered outstanding results1,, for structural, electronic, transport and thermoelectric properties, and a detailed understanding of the fundamental mechanisms which drive the formation of 2DEG in oxide heterostructures.
Starting from our deep knowledge of STO/LAO prototype, future exploration will cover a range of interfaces, films and superlattices of titanates, nickelates, cuprates, and manganites, addressed to the search of 2DEG with the best functionalities in terms of charge confinement, electron conductivity and mobility, field-effect capability.
Currently investigated materials, in synergy with experimental collaborators, include Nb-doped STO superlattices2, Nb-doped and La-doped SrTiO3/SrZrO3 superlattices, LaAlO3/LaNiO3 superlattices. Global aim is tracing a detailed interplay of structural, electronic, dielectric, and functional properties (gas confinement, transport, field-effect behavior). Building a generalized ‘phase diagram’ where fundamental properties and phenomenology are universally related is instrumental to provide seriously predictive guidelines for the experimental work.
Heterostructures hugely widen the basket of viable candidate materials for spintronic technology, far more restricted if only limited to the properties of the corresponding bulk constituents. Heterostructures can benefit of several advantages with respect to the constituent bulk materials: i) built-in planar strain can fine-tune film magnetization and resistivity; ii) spin-orbit coupling (e.g. Rashba effect) induced by inversion symmetry breaking, can be source of large magnetoresistance; iii) magnetization through proximity effects, induced by contact between magnetic and non-magnetic interfaces, enhances spin-injection and spin-diffusion mechanisms; iv) bi-dimensionality, allowing efficient size downscaling and integration in devices (e.g. perpendicular architecture in MRAM).
The accurate theoretical description of magnetic oxide heterostructures and their transport and magnetotransport properties, with inclusion of field and strain effects, is of the outmost importance as complement and guidance for the experimental activity towards the magnetic oxide heterostructures with enhanced spintronic functionalities, such as tunneling (TMR) or giant (GMR) magnetoresistance.
At the functional level we are especially interested in the search for electric-field induced or current-induced switching of magnetization and magnetoresistance, mediated by strain effects, charge and orbital order effects, or spin-orbit coupling. This mechanism can revolutionize the TMR technology and enable a paramount innovation: the electric writing of magnetic bits, and in turn the memory bit size downscaling up to few tens of nm, and the decrease of power consumption from present ~mA switching currents up to currents < 100 mA. Among specific prototypes currently investigated are spin-diffusive ferromagnetic metal/insulating interfaces (Cu2O/LaSrMnO3), spin-valve and tunneling magneto junctions (BaFeO3/SrTiO3/LaAlO3/BaFeO3), magnetostrictive7 (LaNiO3/LaAlO3, LaNiO3/SrTiO3), superconducting (CuO/SrTiO3) and multiferroic (PbTiO3/SrRuO3) superlattices.
Recent years have seen an impressive rise of interest in the field of multiferroics, i.e. systems combining different spontaneous ordering (magnetic, ferroelectric, and ferroelastic) in the same temperature range. In addition to having all the virtues of individual ferroelectric and ferromagnetic materials, these systems potentially hold entirely new capabilities generated by the mutual coupling of these effects, which bring scientists to envision multifunctional devices where this coupling can be effectively controlled and exploited. In electromagnetic materials, the coupling of electric and magnetic polarization can be exploited to enable the electric field control of magnetoresistivity, or vice-versa to tune the dielectric properties with a magnetic field, and can be exploited to project novel devices such as multi-state random access memories, transducers with magnetically modulated piezoelectricity, sensors for the measurements of magnetic/electric fields, modulation of transport properties and capacitance by a magnetic field
As discussed in a series of works by A.F. & N. Spaldin,, the coexistence of magnetism and ferroelectricity in bulk systems is rare. One of the few notorious examples is the hexagonal perovskite YMnO3, brought to the attention of the community by AF & coworkers in 2004. Since then, a variety of oxide families were investigated by AF: the class of insulating orbital-ordered magnetic perovskites (e.g. Pr1-xCaxMnO3) possessing the quality to develop robust magnetoelectricity due to the superposition of colossal-magnetoresistive and colossal-electroresistive effects; the class of layered titanates (e.g. La2Ti2O7), possessing large ferroelectric displacements due to the specific topology, and keen to be doped by magnetic impurities (e.g. Ti-V substitutions) which are demonstrated to interact ferromagnetically; the class of insulating nickelates, e.g. PbNiO3, recently found by AF & coworkers, to be the oxide with the highest known electric polarization (~100 mC/cm2).
The future activity of AF will focus in particular on low-dimensional (layered) oxide systems, including titanates, nickelates, and ferrites. These materials display mechanism of topological (geometrical) ferroelectricity which may be more robust than the ferroelectricity based on cation-anion hybridization acting in conventional ferroelectric perovskites. Furthermore, being geometrically driven, this ferroelectricity is not incompatible with a magnetic sublattice of cations, thus opening perspective of multiferroic behavior, while spin-orbit effects can give rise to magnetic anisotropy and magnetoelectric coupling.
Single-phase bulk systems may have intrinsic limitations concerning multiferroic behavior and/or magnetoelectric coupling. A radical way to overcome this limit is building (designing) heterostructures with the required properties, out of different oxide constituents. The exceptional variety of oxide properties and their epitaxial compatibility allows the possibility to “compose”, in principle, any heterostructure according to the required properties, paving the way for a new generation of memory and logic devices. However, the number of structural, electronic, and magnetic degrees of freedom to be optimized is so large that a ‘trial and error’ experimental procedure is impractical. The advanced First-Principles approach developed by AF & coworkers (VPSIC)3,4 represents a formidable tool to predict heterostructures with the best ferroelectric and magnetic properties, and guide the experiment towards the most promising systems. Out of the countless examples, we can mention two possible magnetoelectric systems:
Piezoelectric/magnetic heterostructures: Superlattices involving interfaces between magnetic and piezo(ferro)electric components (e.g. few mono-layers of magnetic (La,Sr)MnO3 and ferroelectric Pb(Zr,Ti)O3 deposited on an insulating substrate). They have the potential to display strong magnetoelectric coupling as a consequence of proximity effects arising through the interface.
Magnetic/superconducting heterostructures: a typical example YBa2Cu3O7/La1-xSrxMnO3 (YBCO7/LSMO): according to experimental reports, magnetic proximity kills superconductivity in YBCO7 up to a certain length scale. Furthermore, our calculations ,,shows that LSMO can be turned from ferromagnetic to A-type antiferromagnetic if grown on compressive substrates (e.g. LaAlO3, YAlO3) and a c-axis compression is applied. As the c-axis strain switch the axial magnetization off, the proximity effect is suppressed and YBCO7 should recover bulk-like (or film-like) superconductivity. Alternatively to strain, electric and/or magnetic fields can be applied on the manganese side if, in place of a ferromagnetic metal, a magnetic insulator is considered (e.g. Pr1-xCaxMnO3).
Some of the nowadays most exciting areas of scientific and technological advancement in materials science are related to systems characterized by strong electron correlation, i.e. systems where electronic charge tends to be localized in nanometric or sub-nanometric regions of the space, instead of being delocalized throughout the whole crystal, as, e.g. in conventional covalent semiconductors or metals. This localization may give rise to an amazing variety of physical phenomena, such as formation of localized magnetic moments and magnetic ordering, orbital and charge ordering, polaron formation, high-Tc superconductivity, colossal-magnetoresistivity, low-dimensional confinement, Fermi-glass and cluster-spin-glass ordering.
Standard First-Principles Density-Functional theories, so accurate for weakly-correlated systems, usually fail for strong-correlated systems, and theoreticians are called to elaborate novel, above-standard approaches. Implementations accurate enough ad still computationally feasible for large-size systems are very few. As examples we can cite the local-spin density with Hubbard U (LSDA+U), the dynamical mean field theory (DMFT), and the pseudo-self interaction correction (pSIC), originally carried out by AF and N. A. Spaldin in 2003 at the University of California11, and later generalized by AF & coworkers of Trinity College Dublin to variational form (VPSIC)12. The series of excellent results obtained for a range of subjects as diverse as transition metal oxides, multiferroics7,8, diluted magnetic semiconductors,, cuprates,,,, and high-k oxides, testify that the pSIC/VPSIC furnishes a practicable route to first-principles study of strong-correlated materials.
Further extensions and ameliorations of the VPSIC approach are planned concerning methodology (bridging strong-correlated and weakly-correlated regime, matching with Gutzwiller-type double-occupancy Density Functional) and software implementation (efficient parallelization, portability, building of user-friendly interface) eventually leading to license-protected software.
To describe the phenomenology of complex materials, accurate FP methods must be combined (“integrated”) in multi-scale fashion with model (simplified) theories capable to access properties at the nano-, meso-, and macro-scale. This activity is presently cast in different subtopics, corresponding to different projects and collaborative networks:
Electric and thermal transport: the planned strategy is is based on the combination of First-Principles electronic and structural properties with the semiclassical Boltzmann Transport Theory for electrons and phonons. Results for oxides2 shows that the combination of VPSIC and Boltzmann theory deliver accurate electron resistivity, mobility, thermopower, Hall resistivity. Further developments are in order to extend the theory to magnetotransport to linear and above-linear order, spin-Seebeck, thermal conductivity, phonon-drag thermopower. Our final aim is producing a global platform for reliable, multi-scale approach to electric and thermal transport for nanostructures materials, to be applied to design innovative thermoelectric and spintronic devices (MRAM, TMR, GMR). This activity will be carried out in collaboration with a large network of European units, in the framework of a project currently submitted for evaluation to an H2020-NMP call (see the list at page 10). Plans to extend the investigation to ballistic transport have been also established by AF with its collaborator S. Sanvito (Trinity College Dublin), who is the recognized leader of transport modeling in molecular electronics.
Optical and photo-absorption properties: The accurate description of optical, optoelectronic, and photoconversion properties is a formidable challenge, which requires a very accurate description of the electronic properties for nanomaterials whose characteristic size is too large to be described purely by First-Principles. To overcome this limit, A.F. and his coworkers at the CNR-IOM plan to develop a multi-scale framework combining VPSIC and classical molecular dynamics within an optimized divide-and-conquer strategy. To the aim, a project gathering a large network of both European and extra-EU units (including QEERI, Qatar, and REMRSEC, USA) is currently submitted for evaluation to an H2020-NMP call (see page 10).
Magnetic and strong-correlated materials: A vast phenomenology typical of strong-correlated materials is hardly accessible directly by First-Principles calculations. Obvious examples are long-wavelength magnetic, orbital and charge ordering, as well as any type of finite-temperature properties, structural disorder, or any property whose characteristic length-scale is too large to be simulated by computationally heavyweight FP. To overcome this size limit, a popular route is integrating (interfacing) First-Principles and model Hamiltonians. Typical examples are the tight-binding, Hubbard, and Heisenberg Hamiltonians.
AF followed this perspective in the framework of a EU-FP7 Project called ATHENA (“Advanced Theories for functional oxides: new routes to handle the devices of the future“) gathering a European-Indian consortium (CNR of Italy, Trinity College Dublin, the University of Vienna, the JNCASR Institute of Bangalore, the SNBNC Institute of Kolkata, the Harish-Chandra Institute of Allahabad). While the Project produced a number of excellent results.22, more work is planned in collaboration with these groups, to further extend and improve the integration of First-Principles and model Hamiltonians. Final aim is setting a comprehensive approach to describe strong-correlated behavior from bulk systems to devices.
 A. Filippetti, P. Delugas, M. J. Verstraete, I. Pallecchi, A. Gadaleta, D. Marré, D. F. Li, S. Gariglio, and V. Fiorentini, Thermopower in oxide heterostructures: The importance of being multiple-band conductors. Phys. Rev. B 86, 195301 (2012).
 P. Delugas, A. Filippetti, M. J. Verstraete, I. Pallecchi, D. Marré, V. Fiorentini; Doping-induced dimensional crossover and thermopower burst in Nb-doped SrTiO3 superlattices; Phys. ReV. B 88, 045310 (2013).
 A. Filippetti and N. A. Hill, A self-interaction corrected ultrasoft pseudopotential scheme for strongly-correlated and magnetic systems. Phys. Rev. B 67, 125109 (2003).
 A. Filippetti, C. D. Pemmaraju, D. Puggioni, P. Delugas, V. Fiorentini, and S. Sanvito, Introducing a variational pseudo-self-interaction correction approach for solids and molecules. Phys Rev. B 84, 195127 (2011).
 A. Filippetti and A. Mattoni, Hybrid perovskites for photovoltaics: Insights from first principles; Phys. Rev. B 89, 125203 (2014).
 P. Delugas, A. Filippetti, V. Fiorentini, D. Bilc, D. Fontaine, and Ph. Ghosez, Spontaneous electron localization in zero field as the origin of 2-dimensional electron gas at oxide interfaces. Phys. Rev. Lett. 106, 166807 (2011).
 C. Cancellieri, M. L. Reinle-Schmitt, M. Kobayashi, V. N. Strocov, P. R. Willmott, D. Fontaine, Ph. Ghosez, A. Filippetti, P. Delugas, and V. Fiorentini, Doping-dependent band structure of LaAlO3/SrTiO3 interfaces by soft x-ray polarization-controlled resonant angle-resolved photoemission Phys. Rev. B 89, 121412(R) (2014).
 P. Delugas, A. Filippetti, A. Gadaleta, I. Pallecchi, D. Marré, V. Fiorentini; Large band offset as driving force of two-dimensional electron confinement: The case of SrTiO3/SrZrO3 interface; Phys. Rev. B 88, 115304 (2013).
 D. Puggioni, A. Filippetti, V. Fiorentini; Ordering and multiple phase transitions in ultrathin nickelate superlattices Phys. Rev. B 86, 195132 (2012).
 M. Cantoni, D. Petti, S. Brivio, R. Bertacco, I.Pallecchi, D.Marré, G. Colizzi, A. Filippetti, and V. Fiorentini, Band Alignment at Cu2O/La0.67Sr0.33MnO3 interface: a combined experimental-theoretical determination. Appl. Phys. Lett. 97, 032115 (2010).
 F. Ricci, A. Filippetti, And V. Fiorentini, Giant electroresistance and tunable magnetoelectricity in a multiferroic junction; Phys. Rev. B 88, 235416 (2013).
 A. Filippetti and N. Spaldin, On the coexistence of magnetism and ferroelectricity in perovskites, Phys. Rev. B 65 (2002).
 N. Spaldin and A. Filippetti, Why are there any magnetic ferroelectrics? Journal of Magnetism and Magn. Mat. 242-245, 976 (2002).
 B. Van Aken, T. M. Palstra, A. Filippetti, and N. A. Spaldin, The origin of ferroelectricity in magnetoelectric YMnO3, Nature Materials, 3 164 (2004).
 G. Colizzi, A. Filippetti, and V. Fiorentini, Multiferroicity and orbital ordering in Pr0.5Ca0.5MnO3 from First Principles. Phys. Rev. B 82, 140101(R) (2010).
 M. Scarrozza, A. Filippetti, and V. Fiorentini, Ferromagnetism and orbital order in a topological ferroelectric, Phys. Rev. Lett. 109, 217202 (2012).
 X.F. Hao, A. Stroppa, S. Picozzi, A. Filippetti, and C. Franchini, Exceptionally large room-temperature ferroelectric polarization in the novel PbNiO3. Phys. Rev. B 86, 014116 (2012).
 X F Hao, A Stroppa, P Barone, A Filippetti, C Franchini, and S Picozzi; Structural and ferroelectric transitions in magnetic nickelate PbNiO3; New Journal of Physics 16, 015030 (2014).
 G. Colizzi, A. Filippetti, F. Cossu, and V. Fiorentini, Interplay of strain and magnetism of La2/3Sr1/3MnO3 from first principles. Phys. Rev. B 78, 235122 (2008).
 G. Colizzi, A. Filippetti, F. Cossu, and V. Fiorentini, Strain induced magnetic phase transitions in epitaxial (La,Sr)MnO3. Eur. Phys. J. B 70, 343 (2009).
 F. Cossu, U. Schwingenschlogl, G. Colizzi, A. Filippetti, V. Fiorentini; Surface antiferromagnetism and incipient metal-insulator transition in strained manganite films; Phys. Rev. B 87, 214420 (2013).
 T. Archer, C.D. Pemmaraju and S. Sanvito, C. Franchini and J. He, A. Filippetti, P. Delugas, D. Puggioni, V. Fiorentini, R. Tiwari, and P. Majumdar, Exchange interactions and magnetic phases of transition metal oxides: benchmarking advanced ab initio methods. Phys. Rev. B . 84, 115114 (2011).
 A. Filippetti, N. A. Spaldin, and S. Sanvito, Self-interaction effects in (Ga,Mn)As and (Ga,Mn)N, Chem. Phys., 309, 59 (2005).
 A. Filippetti, N. A. Spaldin, and S. Sanvito, Strong correlation and ferromagnetism in (Ga,Mn)As and (Ga,Mn)N, Journal of Magnetism and Magnetic Materials, 290, 1391 (2005).
 A. Filippetti and V. Fiorentini, Magnetic ordering in CuO from first-principles : a cuprate antiferromagnet with fully three-dimensional exchange interactions, Phys. Rev. Lett. 95, 086405 (2005) .
 A. Filippetti and V. Fiorentini, Double-exchange driven metal-insulator transition in Mn-doped CuO. Phys. Rev. B (Rapid Comm.), 74 220401 (2006).
 A. Filippetti and V. Fiorentini, Magnetic ordering under strain and Spin-Pierls dimerization in GeCuO3, Phys. Rev. Lett. 98, 196403 (2007).
 A. Filippetti and V. Fiorentini, Metal-insulating transitions and singlet polarons in one-dimensional Ca2+xY2-xCu5O10. Phys Rev. B 77, 235124 (2008).
 A. Filippetti, G. Lopez, M. Mantega, and V.Fiorentini, Chain metallicity and competition between paramagnetism and antiferromagnetism in underdoped YBa2Cu3O6+x by first-principles calculations. Phys. Rev. B 78, 233103 (2008).
 P. Delugas, V. Fiorentini, A. Filippetti, and G. Pourtois, Cation charge anomalies and high-k dielectric behavior in DyScO3, Phys. Rev. B 75 115126 (2007).
 P. Delugas, V. Fiorentini, A. Filippetti, and G. Pourtois, Conservation of dielectric constant upon amorphization in high-k perovskite oxides, Phys. Rev. B, 76 104112 (2007).
 C. Toher, A. Filippetti, S. Sanvito, and K. Burke, Self-interaction errors in density-functional calculations of electronic transport, Phys. Rev. Lett. 95, 146402 (2005).