[Reading Science] The Divine Technology, Quantum Computers... Spotlight on Korean Semiconductors

Korea Leads in Semiconductor Technology Essential for Qubit Implementation
Challenges Ahead for Commercialization Including Error Correction

Is quantum computing the divine technology that will save humanity? This is the impression after observing the ‘Quantum Korea 2023’ event held from the 26th to the 29th of last month at Dongdaemun DDP in Seoul. The event emphasized that quantum computing could be a means to drastically advance cutting-edge technology and solve crises humanity faces, such as climate change and resource depletion. On the other hand, it was pointed out that quantum computers are not万能solvers and that, unlike existing technologies, numerous challenges remain that may not be resolved in a short time. Let’s explore both the rosy prospects and the shadows of quantum computing.

Digital Bits and Qubits

To understand quantum computers, one must first understand the ‘qubit (Quantum + Bit = Qubit)’. Traditional digital bit computers use digital bits where electricity flowing means 1, and no flow means 0. They perform arithmetic operations such as addition, subtraction, multiplication, and division, and apply these to handle complex calculations.

Qubits apply the quantum science principle of superposition. Quantum science studies the properties of atoms, the smallest units of matter, and holds that until an atom is ‘observed,’ its state is undecided and exists in superposition. This allows for much faster information processing. For example, if there are 3 bits, the possible states range from (0, 0, 0) to (1, 1, 1), totaling 8. Digital bits can only select one of these at a time, but qubits can represent all 8 simultaneously. Increasing this to 20 qubits allows representing 1,048,576 pieces of information at once. Simply put, traditional digital computers operate serially. In a maze with only one path, they try each route one by one, finding the answer through trial and error. Simple tasks take little time, but complex calculations with many variables inevitably take longer. Quantum computers operate in parallel, analyzing all possible paths simultaneously to find the exit. Theoretically, they can perform calculations over 30 trillion times faster than existing supercomputers and over 10 quadrillion times faster than general computers.

Hansang Wook Han, head of the Quantum Information Research Group at the Korea Institute of Science and Technology (KIST), explained, "Qubits simultaneously hold states between 0 and 1, enabling an atom to switch between states A and B to process information. Unlike digital bits that process one at a time, qubits operate simultaneously in numbers, allowing much faster information processing."

The phenomenon of entanglement is also a core principle of quantum information technology. To illustrate, when an atom is split into two, no matter how far apart they are, they maintain an entangled state. If one side’s state changes, the other side’s state changes as well. This principle is applied in quantum sensors that measure minute changes and quantum communication technology that enables unbreakable security.

How Are Qubits Made?

Traditional digital bits are implemented using semiconductors, which are transistors miniaturized and stacked on silicon wafers. To implement qubits, four main technologies are being researched. First, there are superconducting qubits studied by Google, IBM, and others. These are created by cooling materials to absolute zero (?273.15°C), where electrical resistance disappears, and stacking components such as Josephson junctions, capacitors, and inductors on silicon wafers to form qubits.

There is also the ion trap method. IonQ, co-founded by Korean-born Professor Jungsang Kim at Duke University, researches this. Ions (atoms) are trapped in a small vacuum chamber and used as qubits. By adjusting voltage inside the vacuum chamber, ions are suspended in midair and operated like transistors. Instead of making separate devices, each ion (atom) itself is used as a qubit. Photon qubits encode quantum states onto individual particles of light and use them as qubits. The diamond NV center qubit method, researched by KIST, is also gaining attention. In diamond crystals made of carbon atoms, nitrogen atoms replace carbon atoms in certain positions, and adjacent carbon atom sites are left vacant to serve as qubits.

All these qubit implementation technologies require semiconductor process technology. This is obvious for superconducting qubits, which are similar to traditional semiconductor manufacturing. The ion trap method uses semiconductor process technology to create chip-shaped devices that trap ions. Photon qubits require semiconductor processes to create paths for light particles, and the diamond NV center method uses semiconductor processes to implant nitrogen atoms and create vacancies in diamond wafers.

This is where Korea can create momentum to catch up with major quantum powerhouses. Leading quantum computing scholars met at Dongdaemun DDP on the 27th and agreed enthusiastically. Charles Bennett, former IBM researcher; John Martinis, professor at UC Santa Barbara; and Myungsik Kim, professor at Imperial College London, all said, "Korea has astonishing advanced semiconductor process technology. If well utilized, it can become a quantum computing powerhouse."

The Future Created by Quantum Computers

Google’s 50-qubit quantum computer ‘Sycamore,’ developed in October 2019, solved a complex calculation problem that would take a supercomputer 10,000 years in just 200 seconds. What kind of era will this ultra-fast computing power create? The global scientific community calls this the ‘second quantum revolution era.’ Expected breakthroughs include advanced artificial intelligence (AI), revolutionary new drug and material research, solutions to energy and space challenges, unbreakable encryption and network technologies, and ultra-high precision and sensitivity sensors. Taking Korea as an example, quantum computing’s ultra-fast calculation capabilities can optimize semiconductor manufacturing and production, Korea’s representative industry. Enhanced AI and quantum sensor-based autonomous driving technology, ultra-fine process design enabling significantly improved battery performance, and ultra-fast large-capacity data processing with high security will enable revolutionary innovations in future advanced industries such as bio, robotics, and AI. Analyzing DNA composed of billions of base pairs with quantum computers will make innovative disease treatment and new drug development much easier. Ultra-large AI intelligence processing trillions to quadrillions of parameters and human-like humanoid robots will emerge.

Quantum computing will provide clues to solving the mysteries of the birth of the universe, Earth, and life, as well as major issues like resource depletion and asteroid collisions. For example, if quantum simulation reveals the principle of nitrogen fixation, it could drastically reduce the enormous global energy costs of fertilizer production. Humanity’s domain will expand into space, opening an era of planetary exploration, colonization, and energy production. Quantum computing will find revolutionary solutions in all fields, including cryptography, finance, transportation, and power distribution. While it will break all existing encryption systems, new, unbreakable encryption methods will also emerge.

Han said, "Just as the world changed with the advent of computers, the more quantum computers are utilized, the more different the world will be before and after their existence."

There Is Still a Long Way to Go

However, many challenges remain. First, the ‘error phenomenon’ is a problem. Quantum computers operate with extremely small amounts of energy at the atomic level. To implement quantum states (superposition and entanglement), vacuum, superconducting states, and special materials are required. This consumes a lot of energy and resources and is inherently unstable. The cause of quantum computer errors is environmental noise. It arises from environmental factors such as superconducting and vacuum states, as well as hardware materials themselves. Recently, IBM demonstrated error mitigation techniques that pave the way to harness the fast computing speed of actual quantum computers. Google also introduced error correction technology in February.

Early transistors in traditional computers also had many errors, but these disappeared after technological development. Various error mitigation and correction technologies are being researched, and academia is focusing on how to create ‘good qubits’ with fewer errors. For example, if quantum states can be implemented at room temperature and pressure, error-free good qubits can be produced. Developing specialized quantum algorithms for industrial applications is also a major challenge. The reality that ‘primitive computers’ like ENIAC require enormous energy and resources due to superconducting and vacuum states must be overcome. Han said, "Since algorithms based on traditional arithmetic operations are slower on quantum computers, specialized quantum algorithms must be developed. Qubits themselves consume very little energy excluding peripheral devices, so with technological advancement, miniaturization and low energy consumption will be possible."