Samsung uses Honeywell quantum computer to improve battery design

In February 2021, Samsung and Imperial College London launched a new round of cooperation projects, intending to solve the computational problems encountered in the development of batteries through the quantum computer System Model H1 built by Honeywell, in order to improve battery design.

The combination of quantum computer and cloud computing will become the mainstream business model

Because quantum computers are good at specific operations, they are very effective in analyzing huge amounts of data and choosing the best solution. However, when facing general corporate IT computing needs or personal computing needs, quantum computers do not have any advantages over traditional computers. Furthermore, the current technical threshold for manufacturing quantum computers is very high, the suppliers are limited, and the price is quite expensive. It is expected that it will be difficult to enter the market in a few years. Therefore, it is expected that the mainstream business model of quantum computers is to combine quantum computers with cloud computing and provide external services. Computing services.

For example, when there is a need for real-time massive data analysis on the enterprise side, it can be submitted to the cloud computing service provider for quantum computing, and the service provider will then deliver the analysis results and the best solution to the enterprise side; or when the enterprise side submits to the service provider’s computing requirements , The processing by traditional computers may not be ideal, and service providers can also provide additional quantum computing services. The aforementioned Samsung and Imperial College London cooperation project is to use Honeywell’s quantum computer and cloud services to explore battery design solutions.

Preliminary experiment

Samsung and Professor Myungshik Kim, Director of the Department of Theoretical Quantum Information Science, Department of Physics, Imperial College London, and his team Johannes Knolle, Joe Vovrosh, Chris Sai Chris Self, Kiran Khosla and Alistair Smith worked together on early quantum algorithms. The research team created a simulation of the dynamics of an interacting spin model, which is a mathematical model used to examine magnetism.

They conducted experiments on Honeywell’s system model H1, which is Honeywell’s latest generation ion trap quantum computing system


Honeywell Quantum Computer

Honeywell’s H1 has 10 fully connected qubits, can reach a quantum volume of 128, adopts a quantum charge coupled device (QCCD) architecture, and can be quickly upgraded throughout its life cycle. In addition to maintaining the single-qubit gate fidelity ≥99.97% and the two-qubit gate fidelity ≥99.5%, H1 also increases the number of available qubits. The user expects that the interference error of the measurement cross-talker M is 0.2% (current commercial systems) The lowest value measured above), with the characteristics of “intermediate circuit measurement” and “qubit repeated use”.

Break new limits

Complex simulations require H1 to run “deep circuits” and use up to 100 dual-qubit gates to support calculations.

A typical quantum algorithm consists of several single-qubit gates and double-qubit gates. Double-qubit gates, that is, quantum operations between two independent qubits, provide entanglement between qubits, which makes quantum computing more powerful than traditional supercomputers, but these quantum operations are also more difficult to use and costly higher. The complexity of quantum circuits can usually be estimated based on the number of dual-qubit gates.

Professor Kim said that Honeywell’s quantum computer performed well, and the data collected met their expectations for the model-considering the circuit depth required by the algorithm, the data is encouraging

Problem-solving future

Tony Uttley, President of Honeywell Quantum Solutions, said that the project not only benefits the research teams of Samsung and Imperial College London, but also proves that the Honeywell System Model H1 quantum computer can process complex algorithms with high precision, thereby enabling researchers Believing that their results are correct is very important for us to use the Power of quantum computing to solve real-world problems.

Today’s quantum computers are still in the early days of commercial hardware. These systems are often plagued by physical errors that reduce the efficiency of calculations. Honeywell’s H1 uses ion trap qubits to provide extremely high fidelity operation and the ability to retain quantum information for a long time.

The road is long and long, and I will search up and down. Although the field of quantum computing is still developing, it is of strategic significance for companies to begin to explore this field and plan value paths in their industries.

Different manufacturing technologies for quantum computers

The current mainstream quantum processor manufacturing technologies include ion well technology, superconductor technology, semiconductor technology, optical quantum technology, and topological insulator technology.

The advantage of ion well technology lies in the high stability of qubits, but it is not conducive to the manufacture of multiple qubits and the running speed is slow. Honeywell is the pioneer of this technology. The SystemModel H1 quantum computer launched in November 2020 is already equipped 10-qubit processor. Superconductor technology has advantages in manufacturing multiple qubits, but in order to maintain the stability of qubits, the quantum chip must be placed in an absolute zero environment. Google and IBM have adopted this technology.

The advantage of semiconductor technology is that the industry has a high degree of process technology, but it is not conducive to the manufacture of multiple qubits, and the quantum chip must also be kept below absolute zero. Intel is trying this technology. The photon technology does not use quantum chips, but is composed of lasers, photon detectors and lenses. The China University of Science and Technology has adopted this technology. As for the qubits made by topological insulator technology, they are less affected by changes in the external environment and help reduce error correction. The program is still in the development stage, and the main input manufacturer is Microsoft.


The Links:   G150XTN06C PAF600F280-24T