Ferroelectric Memories

Ferrolectric Memories

© Fraunhofer IPMS
FRAM demonstrator with ASIC on PCB.

Hafnium oxide-based ferroelectric memories are among the most promising emerging technologies for future ultra-low power non-volatile memory applications. Therefore, Fraunhofer IPMS’ CNT business unit develops fully CMOS-compatible hafnium oxidebased ferroelectric (FE) devices for integration into a wide range of chip technologies. This innovative lead-free material enables the manufacturing of cost-efficient and power-saving CMOS chips.

 

Our R&D-Services

  • FE-Material research and Memory stack process development
  • Wafer-loop services with Wafer-Fabs for transfer of FE stacks to existing technology nodes
  • Development of innovative memory device and integration concepts
  • Advanced physical and electrical characterization from material level to Megabit memory array
  • Circuit design of memory test-structures and memory arrays
  • System design for in-memory computing solutions

Advantages of Ferroelectric Memories

© Fraunhofer IPMS
Ferroelectric Memories at Fraunhofer IPMS.

Ferroelectric RAM (FeRAM)

  • Hafnium oxide-based FRAM
  • 10+ years storage reliability
  • Fast read/write operation (switching speed in ns range)
  • Ultra-low power & low voltage
  • Array design & technology integration

Ferroelectric FET (FeFET/FeMFET)FRAM

  • Very energy-efficient and non-destructive read
  • Suitable for use in hardware-based neuromorphic computing concepts
  • FEoL (FeFET) and BEoL (FeMFET) compatible integrated gn & technology integration

Ferroelectric storage for low-power systems

Ferroelectric storage at a glance.

Wherever data is processed, this information must often also be filed and stored. The strong development towards networked, intelligent systems requires data processing and storage already in decentralised sensor nodes. Important requirements for this are low energy consumption and at the same time short switching and reaction times. In the business unit Center Nanoelectronic Technologies (CNT) of Fraunhofer IPMS, power-saving, non-volatile memories based on ferroelectric hafnium oxide are being researched and transferred to CMOS-compatible semiconductor manufacturing processes for 200 mm and 300 mm wafer sizes. This enables FeFET- and FRAM-based solutions for front-end and back-end integration, respectively.

In contrast to the previously used perovskite-based materials, hafnium oxide-based memories are CMOS-compatible, lead-free and scalable down to very small technology nodes. As the only non-volatile memory concept, ferroelectric memories are operated purely electrostatically and are therefore particularly power-saving, since only the reloading currents of the capacities have to be expended to write data. In addition to their use as pure data storage, ferroelectric components are also suitable for use in hardware-based neuromorphic computing concepts, where it is important to perform computing operations as close to or directly with the memory cells as possible.

For material development and improvement of the technological processes, there is an integrated FeFET test route at the CNT where various optimisations and developments for the integration of ferroelectric materials can be carried out together with partners and customers. This enables the further development of new concepts as well as the continuous improvement of performance parameters, such as power consumption.

Privacy warning

With the click on the play button an external video from www.youtube.com is loaded and started. Your data is possible transferred and stored to third party. Do not start the video if you disagree. Find more about the youtube privacy statement under the following link: https://policies.google.com/privacy

Further information:

 

Project

3DFerroKI

Hardware-based AI with 3-dimensional ferroelectric memories 

 

Project

FerroSAFE

Field-induced crystallization for robust safety applications

 

Project

Smart IR

AI-based infrared sensors

 

Project

TO.QI

Semiconductor technology modules for quantum computing, AI and the Internet of Things