Although superconducting qubits offer exciting prospects in practical quantum computing, the widely used qubit designs and fabrication techniques currently available do not yet deliver enough performance. In order to enhance the accuracy of quantum computing, a team of researchers from Aalto University, IQM Quantum Computers, and VTT Technical Research Center have found a novel superconducting qubit called the unimon.
It accomplished this by combining the desired characteristics of improved anharmonicity, a complete insensitivity to dc charge noise, decreased sensitivity to flux noise, and a simple structure consisting only of a single Josephson junction in a resonator.
In traditional computers, information is stored and processed as bits that can have one of two potential values: 0 or 1. Qubits, also known as quantum bits, are distinct from conventional bits as a qubit can also exist in a superposition state, or “α|0〉 + β|1〉” in which it can simultaneously have both values. A quantum computer will require at least 10,000 qubits in order to work effectively.
Some quantum processors employ either photonic qubits or trapped-ion qubits. While the former is made up of single photons of light, trapped-ion qubits use charged atoms suspended in an electromagnetic field to store and process information. Another qubit called a superconducting qubit, which has one of the most developed architectural designs, is also used. These qubits are a subset of a larger family of models that make up what we refer to as solid-state quantum computation. Solid-state qubit-based quantum computers don’t have moving components and are built using fabrication methods that were originally created for solid-state conventional computation.
Like any other type of qubit, a solid-state qubit is made by isolating a two-level quantum system. Superconductors and semiconductors have been the main focus of attempts to create solid-state qubits so far. While two semiconductor techniques, quantum dots, and single-donor systems, have shown noteworthy results, the superconducting approach is now the most advanced. As a result, superconducting circuits have become a frontrunner for implementing a scalable quantum computing platform.
A superconducting qubit is actually a circuit loop with an electrical current flowing through it. This circuit consists of metals that, when cooled below a specific critical temperature, turn into superconductors or materials that can carry current without resistance. The current is composed of “Cooper pairs,” a type of electron pairing that happens only in superconductive materials. This is created after a superconducting material hits its critical temperature and the normally repulsive attraction between electrons starts to turn slightly attractive.
Maintaining control over electron-electron interactions, as well as interactions between electrons and other degrees of freedom, is crucial for maintaining coherence. The fact that the electrons condense into Cooper pairs, which then form a single superfluid, gives superconductors an edge in this aspect. Because it requires a specific amount of energy, known as the energy gap, to separate the Cooper pairs, this superfluid can travel through the metal lattice without experiencing any resistance.
Cooper pairs need to transit via the Josephson Junction, which consists of a very thin layer of insulating material (thin layer of aluminum oxide) placed between two layers of superconducting material (layers of aluminum), before they can traverse the voltage gate to the superconducting island. Interestingly, this configuration provides Josephson Junctions with a small degree of self-capacitance, enabling the construction of a charge qubit using only a voltage gate and a Josephson Junction without a separate capacitor. Because of the “Josephson effect,” which occurs under specific circumstances when a non-superconducting material is positioned between two superconductors, Cooper pair electrons can tunnel via the Josephson Junction.
It is important to note that the 0 and 1 states that serve as the foundation for all quantum operations in a charge qubit correspond to the charge states of the island area, i.e., the lack or existence of excess Cooper pairs in the island. As a result, the superconducting island is often referred to as a “single Cooper-pair box.”
Even though superconducting qubits have already achieved quantum supremacy in some computations, current quantum computers still experience errors from noise to the point where their practical applications in fields like physics simulations, optimization, machine learning, and chemistry remain elusive. The complexity of implementable quantum calculations in this so-called noisy intermediate-scale quantum (NISQ) era is primarily constrained by errors in single- and two-qubit quantum gates.
Qubits encoded as charge states are extremely susceptible to charge noise, which also applies to Cooper Pair Box. Cooper Pair Box’s charge noise issue was resolved by creating a qubit with higher-order energy levels, the transmon (transmission-line shunted plasma oscillation qubit).
In theory, quantum error correction might totally eliminate the impact of gate faults. However, error mitigation can only do so much. Due to the two-dimensional architecture of the qubit register and its advantageous fidelity threshold of around 99%, which was previously attained with superconducting transmon qubits in 2014, surface codes are thought to be among the most convincing error correction codes for superconducting qubits. Despite recent significant advancements in the implementation of distance-2–5 surface codes on superconducting quantum processors, it is still necessary to increase the gate and readout fidelities of superconducting qubits, preferably above 99.99%, in order to enable effective quantum error correction with a reasonable qubit count.
Transmon qubits, which can be reliably manufactured and have coherence times up to several hundred microseconds, are currently used in the majority of superconducting multi-qubit processors, resulting in average gate fidelities of 99.98–99.99% for single-qubit gates and 99.8–99.9% for two-qubit gates. The transmon improves upon the original charge qubit’s Cooper pair box by adding a shunt capacitor in parallel with a Josephson junction, which exponentially reduced the sensitivity of its transition frequency to charge noise. However, due to the huge shunt capacitance, the anharmonicity is only 200–300 MHz or 5% of the usual qubit frequency. Due to the necessity to suppress leakage faults to states outside of the computational domain, this reduces the speed of quantum gates that can be implemented with transmons qubits. A high-power readout tone can even excite the transmon to unconfined states beyond the cosine potentia, although this is also restricted by the low anharmonicity of transmon qubits. To speed up qubit operations and enable larger reliabilities within the bounds of the restricted coherence time, a higher anharmonicity is needed. As a result, the team believes finding novel superconducting qubit types that boost the anharmonicity-coherence-time product is important.
Recent years have seen significant advancements to address the above problems via the creation of fluxonium qubits, plasmonium qubits, and quasicharge qubits. However, each of these also has its own disadvantages. The unimon was thus developed by researchers, which consists of a single Josephson junction shunted by a linear inductor and a capacitor in a relatively untapped parameter regime where the inductive energy is mostly canceled by the Josephson energy resulting in high anharmonicity while being fully stable against low-frequency charge noise and partially shielded from flux noise.
By integrating a single Josephson junction into the center conductor of a superconducting coplanar-waveguide (CPW) resonator grounded at both ends, the researchers successfully applied the unimon in a simple superconducting circuit. The junction is inductively shunted since there are no charge islands in the circuit. They claimed that the unimon is the only superconducting qubit with the Josephson junction shunted by a geometric inductance that offers total protection against low-frequency charge noise, aside from the relatively recent fluxonium qubit using a geometric superinductance.
According to researchers, the normal modes of the resonator with a non-zero current across the junction are transformed into anharmonic oscillators that may be employed as qubits because of the non-linearity of the Josephson junction. Since it has the largest anharmonicity, they chose the lowest anharmonic mode in this study as the qubit.
The researchers created devices that included three unimon qubits apiece in order to experimentally demonstrate the unimon. With the exception of Josephson junctions, where the superconducting leads were made of aluminum, they employed niobium as the superconducting material.
The researchers discovered that the unimon qubit could be protected from noise while only requiring a single Josephson connection and having a relatively high anharmonicity. Compared to the junction-array-based superinductors in typical fluxonium or quarton qubits, the geometric inductance of the unimon provides the potential for greater predictability and yield. Most importantly, the team attained fidelities ranging from 99.8%-99.9% for 13-nanoseconds-long single-qubit gates on three different unimon qubits. This is a groundbreaking milestone in quantum computing, bringing us closer to building solid-state superconducting qubits-based quantum computers soon.