Comparing Short and Long-term Reliability of HfO₂ and Al: HfO₂ RRAM Devices

Resistive Random Access Memory (RRAM) devices are emerging as a key technology for next-generation non-volatile memory and neuromorphic computing due to their scalability, low power consumption, and high-speed operation. However, the reliability and stability of these devices remain critical challenges that must be addressed to enable their widespread adoption. This study investigates the reliability and performance of RRAM devices based on HfO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> and Al-doped HfO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> insulators, with a focus on their suitability for advanced computing applications. We performed a comparative analysis of these two technologies, evaluating key factors such as Post-Programming Stability, Fast Drift, Endurance, and Retention. Our results demonstrate that Al-doped HfO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> devices exhibit superior retention of conductance states, with reduced drift and more predictable switching dynamics compared to pure HfO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> devices. Notably, the Al doping technique shows a general trend of reducing the distributions tails in switching voltages, leading to more compact distributions for most conductance levels. These characteristics make AI-doped HfO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> devices a promising candidate for reliable RRAM-based systems in memory and neuromorphic computing applications.