
Cryogenic Technologies
Qubits can be built on several physical platforms, including trapped ions, neutral atoms, photons, and superconducting circuits. Superconducting circuits are one of the leading platforms and are used by many of the largest quantum computing programs.
Why superconducting qubits
Superconducting qubits are electrical circuits built around Josephson junctions, made from metals that lose all electrical resistance at low temperature. Aluminum, the metal most qubit junctions are made of, becomes superconducting at about 1.2 K, but the circuits are operated near 10 millikelvin (about −273.14 °C, colder than outer space) so that thermal energy stays far below the energy of a single qubit excitation: a 5 GHz qubit corresponds to about 0.24 K.
- Fabrication: made with lithographic techniques adapted from semiconductor manufacturing
- Speed: gate operations take tens to hundreds of nanoseconds
- Control: qubits are driven with precisely shaped microwave pulses; in many designs, flux pulses, which change the magnetic flux through a circuit, tune qubit frequencies or couplings
The cryogenic environment
The quantum processor operates inside a dilution refrigerator with several temperature stages, from room temperature (about 300 K) down to around 10 millikelvin. Each stage has limited cooling power, and every signal line that enters the refrigerator carries heat toward the coldest stage. As systems grow, wiring and heat load become a central engineering problem.
The full stack
1. Quantum processor. The chip that holds the qubits and the couplers between them.
2. Control and readout electronics. Control electronics generate the microwave and flux pulses that drive gates. Readout electronics amplify and digitize the weak signals that carry measurement results. In most systems today, pulse generation and digitization sit at room temperature, while the first amplifier stages sit inside the refrigerator. Each qubit is reached through dedicated lines. Control lines going down are attenuated and filtered at the temperature stages to reduce the heat and noise that reach the qubits from warmer stages. Readout signals coming up are too weak to attenuate: they pass through isolators and are amplified, often first by a parametric amplifier at the coldest stage and then by a low-noise amplifier at about 4 K. Quantum gates and circuits are explained in Quantum Computing Basics.
3. Classical interface. Classical computers compile algorithms into circuits, generate control sequences, collect measurement data, and run error correction. Compilers translate circuits for a specific processor; Phi₀’s Qortis™ compiler is in development.
Decoherence
Quantum states are fragile. Interaction with the environment, such as stray electromagnetic fields, thermal noise, or ionizing radiation, causes decoherence, and the quantum information is lost. Coherence times of superconducting qubits have improved steadily and today are typically in the range of tens to hundreds of microseconds.
Quantum error correction protects information by encoding one logical qubit in many physical qubits and checking for errors repeatedly during the computation. Adding qubits improves protection only when the physical error rate is below a threshold set by the code; Google Quantum AI has demonstrated a surface-code memory operating below this threshold (Nature, 2025). This overhead is why large numbers of high-quality physical qubits are needed, and why wiring, heat load, and control at scale become central engineering problems.
Further reading
- M. Kjaergaard et al., Superconducting Qubits: Current State of Play, Annual Review of Condensed Matter Physics 11, 369–395 (2020)
- S. Krinner et al., Engineering cryogenic setups for 100-qubit scale superconducting circuit systems, EPJ Quantum Technology 6, 2 (2019)
- Google Quantum AI and collaborators, Quantum error correction below the surface code threshold, Nature 638, 920–926 (2025)