MEMRIOX Research Topics
Research highlights will be announced here.
Fabrication and characterization of high-quality oxide heterostructures
Metal oxides such as BiFeO3, VO2 or TiO2 , but also SrTiO3 and NiO are typical memristive materials. For layer deposition several techniques, comprising pulsed laser deposition (PLD), atomic layer deposition (ALD), molecular beam epitaxy (MBE) and pulsed vapour deposition (PVD), are provided by the partners. The chemical composition of oxide thin films has been investigated by combined Rutherford Backscattering Spectrometry (RBS) and Proton Induced X-ray Emission (PIXE) channeling and shows the nearly 1:1 transfer for different target compositions and PLD growth parameters. In addition to the above mentioned flexibility concerning the deposited material, PLD is a reliable, versatile, and fast growth process. The oxide ALD facilities of the partner at the TU-BAF will be of vital importance for an atomic-scale control of interface properties in oxide bilayer and multilayer stacks. Here, the known routes for binary oxides will be extended to ternary compounds. For the structural characterization X-ray diffraction (XRD) and transmission electron microscopy (TEM) are employed, with which the structure of other oxide thin films has already been investigated at all partner sites. The generation of shallow and deep defects in oxide thin films will be studied by admittance and deep level transient spectroscopy (DLTS) measurements on the 0.2 mm length scale. The HZDR offers the possibility to connect an MBE chamber to one of the implanters for further ion-beam modification.
Generation of nanostructures from memristive oxide materials
ZnO, TiO2, VO2 and CuO nanowires will be grown as a function of diameter and length. The morphology of the nanostructures will be studied in detail as a function of the growth parameters (pressure, temperature, gradient, catalyst, etc.) and the growth mechanism will be evaluated in order to obtain nanowires with desired diameters. All nanowires will be contacted at both ends for electrical characterization on top of insulating substrates (e.g. SiO2 on Si) using deposition and/or lithography. This easily realized transistor geometry allows the application of a gate voltage to study field induced effects within the nanowires. For comparison, for the same measurement purposes top-down synthesized nanostructures including the electric contacts are created using e-beam lithography.
Ion-beam modification of quasi 1D memristors
In a first step, the contacted oxide nanostructures will be irradiated by noble gases with an ion energy adjusted to the dimensions. Nobel gas atoms by themselves do not induce any electrical effects, and are therefore the ideal species in order to study the electrical effects caused by implantation defects. This experiment will be performed in-situ as a function of the ion fluence, and thus defect level, and will give deeper insides to the memristor mechanism if and how it is defect mediated. Further experiments will be performed for a selected oxide material as a function of the nanostructure diameter in order to determine the influence of the surface. In a second step the same experiments will be repeated under irradiation either with oxygen or the respective transition metal ions. Here, the local stoichiometry and thus the balance between oxygen and transition metal vacancies can be adjusted.
Interface modifications by ion-beam irradiation
As-prepared, high-energy ion implanted memristive films or side-wall isolated nanostructures are contacted by ultra-clean metallic pads by means of molecular beam epitaxy. The near surface vacancy concentration is quantitatively altered by means of low-energy ion irradiation with various ion species, various contact material, and other preparation parameters. The influence of interface defects, which can be hardly controlled by common industry-compatible preparation methods, on the memristive properties will be investigated. Further ex-situ analysis will focus on investigations of the interface structure and stoichiometry and relate those parameters to the memristive effects. In the system SrTiO3 and Al2SiO5 the ordering of structural polyhedra in initially amorphous ALD layers will be investigated.
Electrical characterization of memristive oxides on the nanometer scale
Advanced electrical characterization methods and transport models will be developed to characterize the memristance of thin films with metallic top-top or top-bottom contacts and of single crystals with metallic top-top contacts. Local versus homogenous switching in memristive oxides will be investigated in dependence on the metal-oxide interface and the irradiation-induced effects, respectively. The memristive effect is then evaluated in top-top or top-bottom contact geometry.
Collaboration: F. Kienberger (Agilent Labs, Linz)
Validated memristance measurements will be performed on the micrometer and nanometer length scale. The memristance data will be tested by comparing experimental and predicted time dependent current-voltage measurements on unipolar memristors, bipolar memristors, and on standard resistors.
Collaboration: C. Mayr (TU Dresden)
In-situ investigations of defect formation during ion irradiation by means of positron annihilation spectroscopy
In-situ investigations of these defects during ion irradiation without breaking vacuum conditions or temperature would help to understand, which defects are responsible for the memristive effects observed and how can they be created. Positron annihilation spectroscopy, on the other hand, is a very sensitive tool for determination of the kind, concentration and size of those defects. Therefore, a new apparatus (AIDA) for in-situ investigation of defect formation by means of positron annihilation spectroscopy will be set up.
Contact: K. Potzger (HZDR)
Collaboration: R. Krause-Rehberg (University of Halle)
Comprehensive optical characterization
Classical optical spectroscopy from UV via VIS to IR for thin films grown on top of transparent substrates will be performed, but also for nanowire samples. Thus, the prepared samples of the consortium will be investigated as a function of the growth and ion implantation parameters in order to determine structure-property relations. However, more important is the correlation of the optical data with the memristive properties of the exact same samples in order to corroborate or reject existing models. Furthermore, selected oxide nanomaterials will be spatially resolved investigated at by either cathodoluminescence (CL), µ−photoluminescence (PL), or electroluminescence (EL). Here, the optical properties of single nanoparticles or nanowires can be investigated as a function of shape and diameter, especially for the determination of surface induced effects.
Integration, quality control and reliability assessment
In order to evaluate the applicability of the ion-beam modified memristive devices, a memory array will be designed and structured by means of electron beam lithography, lift-off and etch processes at NaMLab (TUD). The integration of the single elements into larger arrays will reveal their potential susceptibility to influences originating from the specific processing steps. Goal is the minimization of hard failure rates and yield improvement as an important prerequisite for statistical analysis of retention, endurance and reliability. Based on a cross-point structure, the macros will enable real array operation. Electrical characterization of the memristive heterostructures with respect to controlling of full arrays will allow for the assessment of related cross-talk effects. The possibility of optical switching of memristive devices will be evaluated. To this end, the existing spectrometers will be modified in order to enable simultaneous optical and electrical measurements on both thin film as well as nanowire devices.
Computational Material Science
Modeling the structural and electronic properties of the defect-free layers by density-functional methods will yield quantitative data on the surface and interface effects before ion irradiation. These calculations will assist the choice of the suitable oxide material and the metal-oxide material system for the comparison of quasi-1D and layered structures. Furthermore, the thermodynamics of defects and their influence on the memristive properties will be tackled.
Defects in magnetic TiO2 (DETI.2)
Description: Within the project, the magnetic properties of defective TiO2 doped with transition metals will be investigated within a Russia-Helmholtz cooperation. Details about the project can be found here.
Main partners: HZDR and Lomonosow Moscow State University