(Keynote) Nanoscale Memories: What Does Physics Have to Say?

Device scaling and energy consumption during computation has become a matter of strategic importance for future information technologies. The central question addressed in this talk is: What is the smallest volume of matter needed for a memory element? A generic physics-based abstraction for memory devices will be introduced and applied to several baseline memory technologies. The scaling limit of electron-based devices is~5-7 nm due to quantum-mechanical tunneling. Smaller devices can be made, if information-bearing particles are used whose mass is greater than the mass of an electron. Therefore the new principles for devices, scalable to ~1 nm, could be ‘moving atoms’ instead of ‘moving electrons’.

Theoretical feasibility of the 1-nm devices will be justified based on electrical properties of the few-atom systems. Next, thermal properties of atomic contacts and devices such as nanoionic switch/ReRAM memory cell will be discussed to provide an insight whether such small structure could have sufficient thermal stability for reliable operation.

An important task for sub-10 nm nanodevices is recognition of the fundamental role of ‘defects’ that should not be treated as imperfections but instead as controllable entities. The materials nonstoichiometry due to point defects plays a key role in the electronic properties of many materials, for example in metal oxides. Due to these nonstoichiometric defects, the materials often behave as doped semiconductors. In fact, these materials form a special class, sometimes called ‘chemiconductors’.

In addition to solid-state nanoionic devices such as ReRAM) concepts of ‘fluid nanoelectronics’ are proposed that utilize liquid media, which may offer a new promising path to replace the foundation of today’s computing technologies. Examples include nanoionic devices based on electrolyte-filled nanochannels. In principle, such structures might be used to make devices scalable to ~1 nm or below.

Recently there have been suggestions from research that the physical limits of semiconductor technology may in fact be overcome by borrowing from synthetic biology principles. For example, it has recently been shown that DNA can be used to achieve storage densities that cannot be approached by any known solid-state technology. This lecture will examine how one might develop a technology in the semiconductor context that would open up DNA memory to widespread applications.