Research

Hybrid optoelectronic devices

Optoelectronics is all around our daily life e.g. in optical fiber communication, displays, laser pointers, or simply light emitting diodes (LEDs). Organic LEDs are established for many years now while recently, quantum dot LED technology is entering the market. However, many QLEDs use toxic compounds such as cadmium or lead. At the same time, Silicon is the working horse in electronic chip fabrication, but it is barely known that silicon can be used for optoelectronics as well. By shrinking Silicon to nano size, its starts to show photoluminescence. At this point, can silicon nanocrystals (SiNCs) be the material of choice for optoelectronic devices? One of our missions: to engineer devices based on SiNCs to get a non-toxic alternative to today’s QLEDs.

Nanoimprint Lithography and Nanotransfer Printing

Despite streaming and high bandwidth internet connection: Compact discs (CDs) and Blue-rays are an integral part of our lives. The fabrication technology behind is based on molding of polymers, a method also very appealing in devices research. Nanoimprint lithography (NIL) is an unconventional lithographic technique for patterning polymeric nanostructures with high-throughput featuring great precision and potentially low costs. NIL relies on direct mechanical deformation of a resist material by a high fidelity master stamp and can therefore achieve resolutions beyond the limitations encountered in other techniques. Surface plasmonic research, CO2 reduction in electro-photocatalytic cells, optical elements or magnonic crystals to mention few, the imprinting technology provides a powerful toolbox. Even more, when avoiding polymers in the so-called direct Nanotransfer Printing (nanoTP) process, metal contacts can be directly fabricated on sensitive materials like self-assembled monolayers or 2D materials.

Computation by 3D arrangements of nanomagnets

Digital computation by magnetic ordering? That sounds useless or crazy having powerful CMOS technologies available everywhere at low cost. However, if it comes to massively parallel and pipelined digital operations with stringent power constraints, 3D Nanomagnetic Logic might pay off. Based on majority votes that can be re-programmed to the universal NAND or NOR function during run-time, a hybrid co-processor integrated in the back-end-of-line CMOS technology is envisioned. Still far from being ready for mass fabrication, but containing all basic elements for 3D integrated computation with nonvolatile magnetic states.

Magnetic Skyrmions for Logic- and Memory Devices

Skyrmions, i.e. stable spin-textures in ultra-thin ferromagnetic layers, are listed as information carriers (digital states 1 and 0) in integrated, microelectronic and -magnetic devices and systems. In order to use Skyrmions for memory and logic functions, they have to be generated, manipulated and read-out by electronic circuitry. For that, we use finely tuned focused-ion-beam patterning to shape the landscape of potential energy of magnetic multilayer film stacks in order to create, stabilize, annihilate and steer magnetic Skyrmions by means of electrical currents. Our research aims to evaluate a disruptive nanotechnology for the upcoming needs in low-power and ultimately scaled computing and storage systems.

Optically inspired computing devices based on spin waves

As the simple scaling of digital devices comes to physical limits, computing alternatives are heavily researched. One example are computing elements, where properties of matter are directly exploited in order to solve physical equations, perform pattern matching, conduct signal processing or pre-evaluate sensor data. One potential platform to efficiently perform complex computational operations by specialized hardware are so-called spin-waves. Spin-waves in ferro- and ferrimagnetic materials are propagating phase-coherent collective excitations of magnetic spins in form of precessional motion. However, spin-waves do not only propagate with very low losses: They interfere, they add-up, show non-linear behavior and potentially show even power gain and such they can be used as stand-alone computing elements. We try to exploit their properties and purposely manipulate them in order to build devices by adapting optical computing concepts, since spin waves have similar propagation characteristics than optical (electro-magnetic) waves. Our work focuses on Forward Volume Magnetostatic Spin Waves (FVMSW), i.e. the magnetic bias field is normal to the film plane, because isotropic propagation is desirable in spin wave devices.