News

Reaching Towards 2D Memory: Is Magnetic Graphene and Spintronics the Key?

May 14, 2021 by Antonio Anzaldua Jr.

Recent research spurs the prospect of 2D layered materials having the potential to play a significant role in memory and data processing applications.

Recently, more research has been working towards different dimensions for memory––specifically 2.5D high-bandwidth memory and 3D NAND. Aside from those two dimensions, 2D is also making waves with the concept of spintronics and graphene.

 

An example of general 2D, 3D, and 2.5D technology integration.

An example of general 2D, 3D, and 2.5D technology integration. Image used courtesy of Shen et al

 

With research being conducted on 2D memory with spintronics and graphene, the question of how 2D memory and spintronics work together, especially with a focus on data-processing applications. 

 

Spintronics and 2D Memory for Data-Processing?

Industry-standard spintronic devices typically utilize cobalt electrodes to inject and detect spin currents within graphene. Magnetic graphene can inject, detect, and transport spins by itself, leading to the realization of new 2D-spin logic circuitry. 

How does this play into data-processing systems? 

Modern computer systems are built around the Von Neumann architecture, meaning data storage is separate from the processing unit. This separation causes power and time consumption for proposed large data systems, requiring external discrete devices for constant data transmission. 
 
Two-dimensional layered materials provide design flexibility to leave behind standard data processing and storage practices and move into designing high-performance memory devices. These 2D memory devices will have various combinations, from graphene and black phosphorus to hexagonal boron nitride and other metal carbide materials. These combinations lead to a device with a larger bandgap, faster write/read speed, higher memory density, and smaller footprint capability. 

 

Researching Magnetic Moments

Spintronic devices provide high-speed and energy-saving abilities to the table, potential alternatives to conventional electronics. Using a magnetic moment, the magnetic strength and orientation of a magnet, or object that produces a magnetic field, cause the movement of electrons to store and transfer information. Spintronic research has sought out atomically thin materials to scale down spintronic memory devices to meet the current demand for small-scaled electronic footprints. 

The use of graphene material has been a promising approach for 2D-based designs. Still, it has always needed additional components due to the disadvantage of not self-generating spin currents. Studies suggest that if graphene were to be modified to act as a magnetic material, the magnetism would help generate the spin current needed to store and transfer data without external elements. 

Researchers from the Physics of Nanodevices at the University of Groningen, Zernike Institute for Advanced Materials, applied the study on magnetizing graphene to experimental trials to confirm if the modified material would generate a highly spin-polarized current.

 

A simplified schematic of the researched device, which shows the electrical and thermal generation of spin currents in a bilayer graphene/CrSBr heterostructure. Image used courtesy of the University of Groningen

 

The main takeaway from their research is ultimately how, by magnetizing graphene, this material could become a potential choice to further 2D spin-logic devices since it can efficiently convert the charge to spin current and then transfer it over long distances. 

With this potential avenue opening up, the door to further 2D research memory can open up just a bit further. Despite this opportunity, how could 2D materials become integrated into traditional memory devices?

 

Integrating 2D Materials into Traditional Memory Devices

Understanding that graphene plays a crucial role in developing 2D-based devices, researchers at the Chinese University of Hong Kong began experiments to find ways to implement 2D materials with current memory devices to improve performance. 

The researchers proposed a ferroelectric field-effect transistor (FeFET) to be used to showcase resistive switching and charge capture/recapture as it is placed with traditional non-2D material without sacrificing any unique qualities. In theory, combining 2D materials with functional materials can cause higher power consumption and increase the footprint. 

Memristors are used as artificial synaptic devices. These passive, two-terminal electrical components can hold a resistive state that cannot be replicated by any combination of resistors, capacitors, or inductors. The memristor has a variable resistance, which means that it depends on the historical sum of the current that has passed through the device. 

 

A flowchart and schematic that shows the memristive mechanisms for polarization switching and resistive switching through 2D materials.

A flowchart and schematic that shows the memristive mechanisms for polarization switching and resistive switching through 2D materials. Image used courtesy of Li et al

 

There are variations of memristors altered with applied mechanisms; conductive filament (CF), grain boundary (GB), phase-transition, and ferroelectric 2D memristors.  CF memristors can control the ionic movement of particles when metal oxides are injected; however, they suffer from random uncontrolled movement when an electric field is present. 

GB memristors are critical in establishing predefined paths for the migration of electrons, atoms, and ions. The GB memristors were proven to have conducted heat that ruptured the functional material during the experimental trial. 
 
Phase-transition memristors are a very challenging approach; however. During the study, the researchers converted an electric field into the in-plane motion of the pre-intercalated lithium ions. This method was not a replicable approach for mass production. The last tried memristor variation was the ferroelectric approach. 

By applying an external electric field, the material had a spontaneous electric polarization that is reversible. Ferroelectric-based materials offer a more extensive dynamic range due to a sensitive overlay of the channel and electrode. This range will bring forth a new switching power of ferroelectric switching, though further testing is necessary to see if it can outshine resistive switching. 

Even with the waves of innovative research coming out on 2D memory and materials, where will this research lead towards? If more research can be done focusing on 2D, could it leave an impact on the tech industry?

 

Preliminary Stages of 2D Memory

Even with studies being done on graphene and on ways to integrate 2D materials with existing systems, each perspective proves that there is still much to overcome in developing 2D memory devices. Creating 2D memory devices will positively impact the large data technologies, cloud computation, and artificial intelligence markets. 

To meet the large data demand, researchers and developers must continue to dive into the use of 2D materials for new memory devices. Overcoming the design challenges will allow memory devices to establish low power consumption, increased write/read speeds, and smaller device size for 2D memory technology.

 


 

Interested in learning more about other spintronic advancements? Find out more down below.

Researchers Make Strides in “Spintronics” (Spin Electronics)

NYU and IBM Researchers Discover a New Spintronics-Based Fast Data Transfer Solution

Spin-Gapless Semiconductors Show Promise in Spintronic Devices