Author: MemrisTec

Sending electronics into space is no small feat. In Low-Earth Orbit, radiation, temperature extremes, and the vacuum pose challenges that conventional memory simply cannot withstand. The LEOMEM project, part of the DFG Priority Program MemrisTec and funded by the DFG starting in 2025, is addressing these issues. Researchers at TU Munich, University of Rostock, and IHP – Leibniz-Institute for Innovative Microelectronics, led by Prof. Amelie Hagelauer, Prof. Marc Reichenbach, and Prof. Christian Wenger, are developing radiation-resistant RRAM-based memory cells, building on results from the earlier MIMEC project.

First prototype chips, combining enclosed layout transistors (ELTs) with RRAM devices, have already been fabricated at IHP. These cells will undergo multi-stage testing for total ionizing dose (TID), single event effects (SEE), and extreme temperatures. Could memory survive these harsh conditions while remaining energy-efficient? That is one of the key questions LEOMEM seeks to answer.

At the system level, behavioral data from the devices will feed into a design framework to explore secure and adaptive memory architectures. Adaptive error correction codes will work alongside analog control, digital interfaces, and controllers in a fully integrated ASIC prototype. The ultimate goal is a memory system optimized for reliability, energy efficiency, and performance in space missions, paving the way for the next generation of in-memory computing beyond Earth.

How can electronic circuits become flexible enough to wrap around sensors in robotics, wearable medical devices, or smart systems? Researchers from Karlsruhe Institute of Technology (KIT) and Ruprecht-Karls-University Heidelberg, led by Prof.Jasmin Aghassi-Hagmann and Prof.Nima Taherinejad, are exploring this question as part of the DFG Priority Program MemrisTec. The PrintMEM project funded by the DFG since 2025 is advancing printed memristor technology, aiming to bring high-performance in-memory computing directly into flexible and low-power devices.

At the heart of PrintMEM are inorganic memristors made from metal oxides with particle admixtures, carefully designed to reduce variability between components. These tiny devices can switch states in just 100 nanoseconds—several orders of magnitude faster than conventional printed thin-film transistors. But speed is only part of the story. The researchers are investigating multiple circuit architectures—including IMPLY, MAGIG, FELIX OR, and SIXOR—and combining them into functional subunits like sorter circuits.

The project spans the entire development chain, from device fabrication and behavioral modeling to endurance testing and circuit design. By integrating these building blocks into printed logic components, PrintMEM seeks to demonstrate energy-efficient, reliable, and scalable electronics suitable for flexible sensors, robotics, and medical applications. Could this technology redefine how computation is integrated into next-generation devices? By pushing the limits of printable inorganic electronics, PrintMEM is opening a pathway toward highly adaptable and powerful edge-computing systems.

In an era where computing demands are constantly growing, the MuCoRe project is exploring how a single chip could handle multiple hardware configurations simultaneously. Researchers at the University of Rostock and IHP – Leibniz-Institute for Innovative Microelectronics, led by Prof. Marc Reichenbach and Prof. Christian Wenger, are developing multi-context FPGAs (MC-FPGAs) using multi-bit resistive RAM (RRAM) cells as part of the DFG Priority Program MemrisTec. Funded by the DFG starting in 2025, the project aims to combine the speed and flexibility of FPGAs with the efficiency and density of memristive memory.

At the heart of MuCoRe is the multi-level capability of RRAM cells. Each cell can store eight states, supporting up to three hardware configurations per cell and drastically reducing the area and power requirements compared to conventional SRAM-based designs. Specialized analog-to-digital converters (ADCs) provide reliable, energy-efficient readout, and the non-volatile nature of RRAM eliminates the need to maintain inactive configurations.

Beyond individual components, MuCoRe is taking a system-wide approach. The team is using open-source tools and iterative design methods to evaluate critical metrics, including area, power, and performance. By merging analog memristive memory with digital FPGA logic, the project is creating a flexible, energy-efficient computing platform. If successful, MuCoRe could redefine how reconfigurable digital systems are built, providing faster, more adaptable hardware for next-generation applications.

Science knows no gender, and we believe that diverse perspectives are essential to driving innovation and solving the world’s most pressing challenges. MemrisTec is dedicated to create opportunities for scientists from all genders to explore, learn and shape the future of computing.

💡 To all the young women out there dreaming of careers in STEAM: You belong in science, and your contributions will shape the world.

Let’s work together to ensure that every girl knows she can be a scientist, an innovator, and a leader in her field. 🌱✨

The MemrisTec members Ilia Valov and John Paul Strachan (both from FZ Jülich) were choosen by the Nature editors to speak at the 2^nd Nature Conference on Neuromorphic Computing in Shangri-La Beijing, China, from October 13, 2024 – October 16, 2024.

5 years after the 1^st conference they will discuss in an interdisciplinary audience how to create more efficient and intelligent computing systems. Among the further speakers is also one member of the MemrisTec International Advisory Committee, Wei Lu from the University of Michigan, USA.