MIT Discovers Quantum-System-on-Chip to Control Qubits

June 10, 2024 by Lisa Boneta.

Using a modular fabrication process, the team created a quantum-system-on-chip that integrates artificial atom qubits onto a semiconductor chip.

Researchers at the Massachusetts Institute of Technology (MIT) and Mitre Corporation recently demonstrated a scalable, modular hardware platform that integrates thousands of interconnected qubits onto a customized integrated circuit.


Quantum computer

Researchers believe their microchiplets based on the color centers of diamonds may help enable practical quantum computing. Image used courtesy of Negro Elkha via Adobe Stock license


This quantum-system-on-chip (QSoC) can efficiently control a large array of qubits, making another step forward in the march toward widespread quantum computing.


Tuning Qubits With a Quantum SoC

One of the main challenges of working with qubits, the building blocks of quantum systems, is that they are fragile and susceptible to errors, making them notoriously hard to control.

At MIT, researchers have introduced a new quantum-system-on-chip (QSoC) architecture to meet the demands of controlling, tuning, and scaling dense arrays of qubits. To scale quantum systems, multiple chips can be connected with optical networks to create larger quantum communication networks. 


Diagram of the system architecture

Diagram of the system architecture, which includes both an optical interface and QSoC. Image used courtesy of ArXiv

At the core of the QSoC architecture is an “entanglement multiplexing” protocol, which allowed the researchers to tune qubits over 11 different frequency channels. The QSoC module itself contains a CMOS application-specific integrated circuit (ASIC). The ASIC provided a voltage bias to tune the frequency of the qubits' electronic spin to a predefined set of frequencies. 

This scalable and integrated design enabled the researchers to manage and coordinate thousands of qubits—solving a critical piece of the quantum computing puzzle. 


Qubits From Diamonds

In quantum computing, color centers in diamonds can be used as “artificial atoms.” Diamond color centers are compact, solid-state systems with long coherence times. This means that the qubits can remain stable for a longer amount of time because of the cleanliness of a diamond’s environment. Qubits also have photonic interfaces that allow them to be entangled with non-adjacent qubits, enhancing scalability. 

Researchers at MIT have used the spectral frequency of diamond color centers to communicate with each individual atom by voltage-tuning them. To surmount the challenge of communicating across thousands of qubits, the team integrated a large number of diamond color centers on a CMOS chip to create the dials to tune the qubits accordingly. 


Fabrication Process

The researchers needed specialized hardware to build such a QSoC. They fabricated an array of diamond color center microchiplets from a block of diamond. They then post-processed a CMOS chip to add microscale sockets matching the diamond microchiplet array. Finally, the team used an in-house setup to apply a lock-and-release process to transfer the microchiplets into the sockets on the CMOS chip.


The lock-and-release integration process to transfer the quantum microchiplet array

The lock-and-release integration process to transfer the quantum microchiplet array. Image used courtesy of ArXiv

Linsen Li, an electrical engineering and computer science (EECS) graduate student leading the QSoC research, says that the team has “iterated and developed the recipe to fabricate these diamond nanostructures in an MIT cleanroom.” Developed over several years, this recipe includes 19 steps of nanofabrication to yield the diamond quantum microchiplets.


Road to Commercialization

In addition to building a QSoC, the researchers developed an approach to characterize and scale the system. They built a custom cryo-optical metrology setup to tune a chip with 4,000 qubits while maintaining its spin and optical properties. While the QSoC offers to make quantum computing a practical reality, researchers will need to refine the materials to make qubits or develop more precise control processes.