 |
| Memory devices based on a-COx. Credit: Nature Comms |
Memory that constitutes resistance as a state variable encompasses a broad range of materials and switching mechanisms
Of these memory technologies, some, namely magnetic random access memory (MRAM), phase-change memory (PCM) and reduction/oxidation (redox) memories, have received more attention from the scientific community and the semiconductor industry and are thus in a more advanced state of research and/or development.
At IBM Research – Zurich, we are pursuing yet another resistive memory concept based on carbon. Carbon-based memory could be a significant complement to the rapid advances in carbon-based nano-electronics. This could pave the way for potential all-carbon computing devices of the future.
The elemental nature of carbon would enable a carbon-based memory to be scaled down to very small feature sizes and to be immune to compositional changes that typically plague alternate multi-elemental non-volatile memory materials.
In addition, the high resilience of carbon to a variety of external stimuli would ensure robustness and endurance of such a carbon-based memory.
Today, a team of scientists based in Switzerland published a significant milestone in the prestigious peer-review journal Nature Communications on the development of carbon memory. I interviewed three of the paper's authors to gain some further insight on their research.
 |
| Claudia Santini |
You compare oxygenated amorphous carbon with graphite oxide in the opening abstract
of the paper. Why?
Claudia Santini: Graphite
and graphene oxide have been widely investigated in the past years for resistive
memory applications.
However, their performance fell short of the expectations and, moreover, the fabrication methods of these materials are somehow unreliable and
do not allow properly controlling their properties on wafer-scale.
This is a
clear disadvantage for electronics applications. On the other hand, in our work
we demonstrate that oxygenated amorphous carbon, which is a material made of
carbon and oxygen like graphite/graphene oxide, can be produced by a much
simpler method, its properties can be controlled on a wafer scale and
outperforms graphite/graphene oxide for resistive memory applications.
What inspired the use of oxygenated amorphous carbon?
CS: A relatively simple idea. Amorphous carbon memories have been investigated in the
latest years by some research groups including the group working with Evangelos Eleftheriou at IBM. These memories have always shown limited endurance due to the difficulty
to break the conductive carbon filaments that are thought to be responsible for
the low resistance state.
As it is known that carbon-based materials break down
by Joule heating at much lower current densities when exposed to oxygen, I
thought that the addition of oxygen as a dopant to carbon-based memories could
facilitate the breaking of the carbon filaments and improve the memory
endurance.
This was correct and we were able to demonstrate endurance higher
than 104 cycles without current control devices. Of course, the
switching mechanism we finally demonstrated was more complex than our initial
idea, but the general concept was correct.
 |
Comparison of the fabrication methods of GO and a-COx.
Credit: Nature Communications |
Abu Sebastian: Both
of them are non-volatile memory (NVM) candidates. In both cases we use resistance as
the state variable to store information.
Projected
memory is a radical rethink of a phase change memory device where we mitigate
or eliminated several drawbacks of conventional phase change memory technology.
PCM
is arguably the most advanced NVM technology and projected PCM will give
significant impetus to that. Compared to
phase change memory, carbon-based memory is at its infancy. But, carbon-based
memory has some unique advantages.
For example, it uses a very common element as
the memory material, could possibly scale to very small dimensions, can
integrate seamlessly with other carbon-based electronic devices such as logic
devices and interconnects. Hence in some applications, carbon-based memory and
oxygenated carbon memory in particular could be much better suited.
What
was the biggest challenge?
CS: It is not obvious to mix carbon and oxygen and make a memory
out of it, which needs to withstand thousands switching cycles and it took some time before being able to convince ourselves that this could work.
But the biggest technical challenges were to produce the material and assess the reproducibility of the process and to
study and understand the switching mechanism in oxygenated amorphous carbon
memories.
Generally, the switching mechanism in resistive memories is an object of
a long debate. The difficulty to study this mechanism comes from the fact that
the switching occurs at sub-nanometer scale and it is very challenging to visualize and
analyze. Thanks to a careful electrical characterization and a detailed
microscopy and spectroscopy analysis we have been able to shed some light on
it.
Wabe Koelmans: Inherent variability in resistive switching devices
(cell-to-cell and cycle-to-cycle).
When do you think carbon-based electronics will enter the market?
WK: There is a company called Nantero working towards a commercial memory product based on carbon as a switching material. It is unclear what
their timeline is for a commercial product. As of now, there are no commercial carbon-based memory chips available.
CS: Carbon-based electronics has driven a lot of interest in research in the latest years. Several
electronics devices based on carbon allotropes have been demonstrated:
interconnects, transistors, memories. As it was stated about one year ago by
one IBM colleague, Wilfried Haensch, according to the semiconductor industry’s
roadmap, transistors will need to have features as small as 5 nm already in
2020 and this is when carbon is expected and will have to replace traditional
silicon electronics.
Claudia, you recently joined a new company. Are you continuing this research?
CS: Not exactly. Optotune is a company working in the field of deformable optics so we investigate new polymer and carbon based materials for optical rather than electronics applications. Very soon we will start working on a new challenging and interdisciplinary project where an interesting mixture of MEMS, polymer and biomedical technologies will be developed to launch a new product.
Based on carbon? It can be, but you will know more about it in the future.
Oxygenated amorphous carbon for resistive
memory applications, Claudia A. Santini, Abu Sebastian, Chiara Marchiori, Vara Prasad Jonnalagadda, Laurent Dellmann,
Wabe W. Koelmans, Marta D. Rossell, Christophe P. Rossel& Evangelos Eleftheriou, Nature Communications, 6, Article number: 8600,doi:10.1038/ncomms9600
Labels: Carbon, IBM Research - Zurich, memory, nature communications
