{"id":1521,"date":"2026-01-12T11:26:51","date_gmt":"2026-01-12T16:26:51","guid":{"rendered":"https:\/\/faculty.eng.ufl.edu\/quanta\/research\/ferroelectric-alscn-for-in-memory-computing-applications\/"},"modified":"2026-03-20T13:33:10","modified_gmt":"2026-03-20T18:33:10","slug":"ferroelectric-alscn-for-in-memory-computing-applications","status":"publish","type":"page","link":"https:\/\/faculty.eng.ufl.edu\/quanta\/research\/ferroelectric-alscn-for-in-memory-computing-applications\/","title":{"rendered":"Ferroelectric AlScN for In-Memory Computing Applications"},"content":{"rendered":"\n
<\/div>\n\n\n\n
\n

Overview<\/strong><\/p>\n\n\n\n

\n\n\n\n

The conventional von Neumann architecture has intrinsic limitations such as memory bottlenecks and excessive energy consumption, which restrict further advances in high-performance and energy-efficient computing. To overcome these challenges, in-memory computing has emerged as a promising paradigm that eliminates the data migration between memory and processing units.<\/p>\n\n\n\n

A key requirement for such architecture is the use of non-volatile memory (NVM) devices capable of stable switching and analog programmability. Among them, ferroelectric memory devices have attracted particular interest owing to their unique superiorities with respect to power consumption, operation speed and endurance. However, conventional ferroelectric materials such as lead zirconate titanate (Pb(Zr,Ti)O3<\/sub>) and hafnium zirconium oxide (Hf(Zr)O\u2082) often face challenges such as reliability issues and performance limitations, which have motivated the search for alternative ferroelectric materials. <\/p>\n\n\n\n

Aluminum scandium nitride (AlScN) stands out as a scalable ferroelectric material that combines robust polarization behavior with back-end-of-line process compatibility. Firstly, we investigate the fundamental aspects of AlScN, including its growth methods, structural characteristics, and key material properties. Based on the understanding of material physics, we design and fabricate ferroelectric memory devices. Using the AlScN-based capacitors and field-effect transistors, we analyze their ferroelectric properties through measurement of P-E<\/em> hysteresis loop, C-V<\/em> and I-V<\/em> characteristics. Following the device characterizations, we further design a crossbar array architecture and integrate it with peripheral circuitry for in-memory computing applications. This systematic approach effectively bridges each field of materials, devices, and circuit systems.  <\/p>\n<\/div><\/div>\n\n\n\n

<\/div>\n\n\n\n

Featured Publications:<\/h3>\n\n\n\n
<\/div>\n\n\n\n
\n
\n
\n
\"\"<\/figure>\n<\/div>\n\n\n

Aluminum Scandium Nitride as a Functional Material at 1000 \u00b0C<\/a><\/span><\/strong><\/p>\n\n\n\n

This paper presents a comprehensive study of the dielectric, piezoelectric, and ferroelectric properties of AlScN thin \ufb01lms using TaSi2<\/sub>\/AlScN\/TaSi2<\/sub> capacitors in extreme thermal environments, demonstrating functional stability up to 1000 \u00b0C, which highlights the material\u2019s potential for high-temperature electronics such as aerospace, hypersonics, and nuclear reactor systems.<\/p>\n<\/div><\/div>\n<\/div>\n\n\n\n

\n
\n
\"\"<\/figure>\n<\/div>\n\n\n

An Efficient Device Model for Ferroelectric Thin-Film Transistors<\/a><\/span><\/strong><\/p>\n\n\n\n

This paper describes an efficient model for Fe-TFTs with a small set of parameters, validated against experimental I\u2013V<\/em> characteristics. The model integrates stochastic multi-domain ferroelectric switching with TFT electrostatics and carrier transport using a virtual source approach, predicting a large memory window and demonstrating high-speed, power-efficient MAC operation in crossbar array simulations.<\/p>\n<\/div><\/div>\n<\/div>\n<\/div>\n\n\n\n

References:<\/h2>\n\n\n\n