In pursuit of more efficient energy computing new devices (New Devices for More Efficient Energy Computing) developed at UC Santa Barbara promise improvements in design and data storage.
Researchers from the laboratory of Kaustav Banerjee, Professor of Electrical and Computer Engineering, have published a new paper describing many of these devices, “2D Materials-Based Quantum Engineering Devices for Next-Generation Information Processing and Storage “, in the journal Advanced Materials. Arnab Pal, who recently received his doctorate, is the author.
Each device is designed to solve traditional computing challenges in a new way. All four operate at low voltage and have very low leakage, unlike the metal-oxide semiconductor field-effect transistors (MOSFETs) found in smartphones that draw power even when they are ‘off’ away. But because they are based on a design process similar to that used to make MOSFETs, the new devices can be manufactured at scale using existing standard manufacturing processes for semiconductors.
The most promising of the two information-processing devices, according to Banerjee, is the field-effect transistor, or spin-FET, which uses the magnetic moment – or spin – provided by electrons, physical device. In this case, the elements are in the group of compounds of transition metal dichalcogenides, based on transition metals.
Unlike spin-FETs, charge field-effect transistors, or TFETs, work by taking advantage of quantum mechanics. Because of what is known as wave function input, electrons are able to pass through a critical electrical barrier instead of passing through it, as is the case with traditional transistors. TFETs can also operate at lower voltages, consuming less power and generating less heat. TFETs made of 2D material are efficient due to their thin and manageable electron tunneling barrier, which improves electron flow and allows the device to operate efficiently.
In order to keep data private, the hard drive of the device must be configured so that the memory is saved even when the power is turned off. To do this with a normal MOSFET, Pal said, “You have a source (electrons) and a drain (where the electrons collect) and a channel between them, which controls the current of the carriers. The channel also works as a barrier so that the electrons cannot pass through – out – or conduct well enough to pass through, ie in the state. To generate flux in the device in a normal fashion, one must apply a positive gate bias – a voltage – that attracts negative electrons from the source through the channel.
Meanwhile, floating-gate field-effect transistor (FG-FET) memory devices work on a similar principle to MOSFETs, but instead of having one electrode grid, there are two. The embedded electrode is called a floating gate. When the MOSFET is not activated, there is no further charge on the floating gate.
However, during each program cycle, a very high voltage is applied to the gate, causing many carriers (electrons) to be released from the channel and placed in a trap. The accumulated negative gate charge makes it difficult to turn on the device, thus causing it to turn off.
There are challenges to this process, according to researchers. One, Pal said, “pulls a lot of weight on a floating door, which requires a lot of power.” In addition, Banerjee added, “In order for these electrons to enter the floating gate to be used in the device, they have to pass through the dielectric layer. And once there, they can to bail out, creating debt retention problems.
A variety of fee-based and non-fee-based applications that use 2D objects can enable next-generation information processing and security technologies.
Another problem is that although many FG-FETs are nearby – and, Pal said, “We want to put as many as possible on the chip, for example, to increase the capacity of the USB key” – the charge stored in the devices. interacting, affecting neighboring devices. The use of critical 2D objects minimizes this interaction while increasing the control of each device, thus delivering high performance even when multiple devices are crowded with chips.
Another way to store information is to use magnetic tunnel junctions (MTJ), which use electrons to store data. An MTJ consists of two magnetic layers separated by a thin insulating layer; the orientation of the magnetic moment of the fabric determines the resistance of the device. However, like FG-FET, MTJ faces challenges related to power consumption, stability, and scalability. In addition, ultra-thin 2D material provides a possible solution by reducing the interaction between neighboring MTJs, thus improving data storage quality and quality.
Although devices based on 2D materials can bring improvements in power and local efficiency compared to those made with solid materials, exceptional improvements can be achieved only by changing the design. Enter the radical new architecture called quantum computing. Based on the density of quantum bits, called qubits, quantum computers create short circuits to produce large performance gains in speed and efficiency for certain computing tasks.
The unique properties of the electronic properties of 2D objects that enable good performance for other charge qubits also lead to many other types of well-designed qubits, called spin-, valley -, and spin-valley qubits.
In a spin qubit, the ground property (on or off) is defined by the electron, or quantum state, of the qubit, which is on or off. The state is changed by turning the spin, up and down or vice versa, to change the state of the qubit and enable the qubit to function.
A valley qubit works a bit differently. Its position is determined by the energy of the electron rather than its extension. The change in the speed of the electrons causes a change in the state of the qubit, helping to understand the function of the qubit.
Finally, the properties of 2D objects can also help to understand a third type of qubit, the spin valley qubit, whose state is defined by both the energy and spin of the electrons. Because spin-valley qubits couple these two degrees of freedom, they are believed to be able to resist entanglement, thus allowing long-term and complex quantum computations to be performed before merging together.
“The emerging devices, which are supported by the unique properties of 2D materials, offer the promise of high-performance power and safety,” said Pal, “making the contribution more than Moore and Stimulating new discoveries in solid state physics and their applications.”
Source: UC Santa Barbara
