A team of researchers from MITP’s Center of Shared Research Facilities has found a way to control oxygen concentration in tantalum oxide films that are produced by atomic layer deposition. The films could be used as the basis for new forms of nonvolatile memory.
Data storage and processing solutions are very central to modern technology. Because of this, research teams and companies are always trying to find new types of computer memory. A major goal for the computer industry is to develop universal memory. A storage medium like this would combine high speed of RAM with a nonvolatility flash drive.
A new technology that will be integral to creating a memory like this is resistive switching memory or ReRAM. ReRAM changes resistance across a memory cell as a result of applied voltage. Each cell has a high and a low resistance state, which can be used to store information.
A ReRAM cell can be a metal-dielectric-metal structure. Oxides of transition metals have proved useful as a dielectric component of the layered structure. When the voltage is applied to a memory cell based on a transition metal causes oxygen migration, which changes the resistance and makes the distribution of oxygen concentration in the film a crucial parameter that determines the functional properties of a memory cell.
Despite advances in ReRAM development, flash memory has no sign of losing ground. Flash memory allows for three-dimensional memory cell stacking. This enables a greater storage density. In contrast, oxygen-deficient film disposition techniques that are normally used in ReRAM design aren’t applicable to 3-D architectures.
MIPT researchers sought out to change this. They concentrated on atomic layer deposition, a chemical process that produces films on the surface of a material. ALD has become increasingly widespread in the last ten years, with many applications in nanoelectronics, optics and biomedical industry. There are two huge advantages to atomic layer disposition. The first is an unprecedented control over film thickness. Depositing films are several nanometers thick, that have an error of a fraction of a nanometer. The second advantage is that ALD allows for conformal coating of 3-D structures. This is problematic for most of the nanofilm deposition techniques currently in use.
During the ALD process, a substrate is exposed to two chemicals known as the precursor and the reactant. The chemical reaction between these two substances produces a coating layer. There are also compounds called ligands. Ligands facilitate the reaction. But in an ideal ALD process, they must be completely removed from the resulting film when the interaction with the other chemical has occurred. Choosing the right substances for this process is very important. It is difficult to prove oxide films with variable oxygen concentration by ALD and essential for ReRAM.
"The hardest part in depositing oxygen-deficient films was finding the right reactants that would make it possible to both eliminate the ligands contained in the metallic precursor and control oxygen content in the resulting coating," said Andrey Markeev, who holds a Ph.D. in physics and mathematics and is a leading researcher at MIPT. "We achieved this by using a tantalum precursor, which by itself contains oxygen and a reactant in the form of plasma-activated hydrogen."
Confirming the findings was difficult. When the sample is removed from the vacuum chamber, which is what holds it during ALD, and exposed to atmosphere, modifications in the top layer of the dielectric happen. This makes it close to impossible to detect oxygen deficiency with analytic techniques that target the surface sample.
"In this study, we needed not just to obtain the films containing different amounts of oxygen but also to confirm this experimentally," said Konstantin Egorov, a Ph.D. student at MIPT. "To do this, our team worked with a unique experimental cluster, which allowed us to grow films and study them without breaking the vacuum."
This study was published in ACS Applied Materials & Interfaces.