Trajectories in quantum computing research often center around the fundamental building block: the qubit. In the realm of superconducting qubits, fluxonium has emerged as a compelling platform. It consists of three elements in parallel: a capacitor, a Josephson junction, and a superinductor. The qubit has three remarkable properties at half-flux quantum bias: (1) It has high coherence without advanced material engineering thanks to the mitigated dielectric loss at low frequency, the protection from quasiparticles due to symmetry, and the protection from high-order flux noise due to the large inductance. (2) It has large anharmonicity, which leads to excellent scalability/extensibility. (3) It has intricate selection rules that kindle transformative control/readout techniques.

As researchers continue to refine fabrication processes and harness new materials, fluxonium qubits are poised to exceed the performance barriers that have limited other superconducting platforms. By combining outstanding coherence with advanced architectural flexibility, fluxonium not only promises to tackle complex computational tasks but also opens up unprecedented avenues for sophisticated control and readout. This dual advantage—in performance and design—has the power to reshape how we build and scale quantum processors, offering a radical leap forward in both academic research and commercial applications.

References:
Physical Review X 9, (041041), Viewpoint, News

PRX Quantum 3 (3) 037001, Press Release, News, Tweet

Physical Review X 11, (021026)

Physical Review Research 4 (023040)

Nature Communications 12 (6383)

My PhD thesis, selected for a distinguished dissertation award

To harness the power of the quantum world, we must not only encode the information but also orchestrate the interactions between otherwise isolated quantum systems, which requires tailoring the effective Hamiltonian of the system. In superconducting qubits, in-situ Hamiltonian engineering can be done by varying the system parameters with external magnetic flux, or by dressing the qubits with microwave drives. The former technique inevitably opens up a decoherence channel to environmental flux noise, while the latter often entails spurious effects.

 
Recently, the Floquet framework has emerged as a powerful tool to analyze and elucidate the electrodynamics of strongly driven quantum systems. Motivated by a prior theoretical study, we set out to explore the physics of a strongly driven quantum processor, implementing the XXZ Heisenberg interaction model that has adjustable spin-exchange (XY) and spin-spin (ZZ) couplings. We demonstrate a suite of high-fidelity qubit gates as applications of the protocol, including an extensible CCZ unitary. This sets the stage for high-performance quantum simulation and quantum information processing using high-coherence fixed-frequency solid-state qubits.

References:

Nature Physics 20 (1), Tweet, Behind the paper

The superiority of quantum information processors over their classical counterparts lies in their entangling power: the bigger the entangled state we can efficiently generate, the more capable the platform becomes. Yet, most of the quantum operations in the existing toolboxes arise from simple two-body interactions. The mainstream solution to generate large-scale entanglement is to compile circuits using only two-qubit gates, which comes at the cost of increasing circuit depths.
The alternative and elegant approach is to instead program a native multi-qubit operation. We achieved this by first utilizing the higher qubit levels for shelving a certain computational state, thereby realizing a three-qubit CCZ/Toffoli gate (F=96.18%) that can be extended to N qubits. In another innovative protocol, we leveraged non-commuting operations to engineer a single-step iToffoli gate (F=98.26%), then used it to compute frequency-domain molecular response properties. To further enhance the capabilities of superconducting quantum systems, we are continuously exploring potential improvements of these protocols, new methods for multipartite operations, and their meaningful applications.

References:

Nature Physics 18 (7), Press Release, Tweet

arXiv:2302.04271

Nature Physics 20 (1), Tweet, Behind the paper

Nature Communications 15 (7117)

Conventional wisdom dictates that the concept of a classical bit serves as the foundation for its quantum counterpart. While the classical bit can exist in the 0 or 1 state, a qubit exhibits the remarkable property of being able to exist in a superposition of these states, spanning a Hilbert space of dimension 2. Composite quantum systems are built by combining multiple qubits, thus expanding the Hilbert space, enabling the storage and processing of intricate quantum information.

However, the challenge of enhancing the capabilities of quantum devices by scaling up to a significant number of qubits is a formidable one. Instead, many quantum systems are composed of multiple energy levels, making them more intuitively suited to function as qudits. Qudits can be employed for implementing quantum error correction codes with higher error thresholds, simulating quantum systems with larger dimensions, or engineering protocols that were otherwise unattainable. My focus lies in discovering efficient techniques for performing multi-qudit entangling operations and harnessing higher eigen-levels to create novel methods that are otherwise not possible with qubits.

References:

Nature Communications 12 (6383)

Nature Communications 13 (7481)

Nature Physics 20 (1), Tweet, Behind the paper

Nature Communications 15 (7117)

Quantum information possesses an inherent fragility. Even minor disruptions from the surrounding environment can disrupt quantum coherences, effectively transforming the system from a quantum state to a classical one. To harness the complete potential of a universal quantum computer, we must operate it within the fault-tolerant regime. This necessitates the encoding of logical qubits across a substantial number of physical qubits, following specific quantum error correction (QEC) schemes.It's worth noting that in cases where the physical noise exhibits a structured pattern with a single dominant error channel to be corrected, certain QEC codes offer more efficient solutions for implementing fault-tolerant systems.

One promising platform to implement biased-noise architectures is the Kerr-cat qubit. Recently, our team at Berkeley successfully constructed a planar architecture based on the Kerr nonlinear elements, demonstrating its scalability. We developed tailored benchmarking capabilities to quantify the errors and bias of the operations, validating the analytically predicted performance of the cats. Notably, at a specific operating regime, we observe gate fidelities crossing the theoretical fault-tolerant threshold.

Heavy fluxonium qubits biased at certain external flux values also operate in this regime, with the energy relaxation prohibited by the disjoint computational wave functions while the pure dephasing is limited by 1/f flux noise. While the computational subspace has zero dipole moment, higher levels can be utilized to implement entangling gates with high fidelity. Importantly, the non-Markovian nature of the noise should allow high-fidelity gates, streamlining a novel pathway toward fault-tolerant quantum computation.

References:

Physical Review Letters 120 (150503)

Physical Review X 11, (021026)

Physical Review Research 4 (023040)

Physical Review X 14 (041049)

arXiv2411.04442

Decoherence errors arising from noisy environments significantly hinder progress in quantum computing and quantum information processing. Performing quantum error correction (QEC) using the Gottesman-Kitaev-Preskill (GKP) protocol has been shown to offer a promising solution to this problem in trapped ions and superconducting circuits. In contrast to active QEC, a quantum system can be constructed such that its Hamiltonian corresponds to stabilizers, offering passive protection from errors analogous to topological protection in quantum spin models. Drawing parallels to the GKP encoding scheme, it is conceptually promising to construct a qubit with eigenstates forming protected grid states, which has thus far remained elusive.

We have been working on the implementation of a scalable superconducting qubit with grid-like eigenstates by integrating an effective Cooper-quartet junction with the quantum phase-slip element embedded in a high-impedance circuit. The spectroscopy results reveal a doubly degenerate states separated from each other by large energy gaps, matching our multimodal theoretical framework. Our findings showcase the versatility of the superconducting circuit toolbox, setting the stage for future exploration of advanced superconducting artificial atoms with emergent properties and nontrivial topological signatures.

References:

Nature 585 (9)

Physical Review X 12, (021002)

In prep.

Solid-state quantum technologies are poised to transform computing, sensing, and secure communications, yet their practical success hinges on innovations in the research and development of materials. By cross-pollinating quantum device design and measurements with advances in novel materials, researchers can create qubits and circuit elements that are more stable, scalable, and efficient. Nowhere is this synergy more evident than in the engineering of new materials for Josephson devices, the building blocks of superconducting qubits. These devices, which exploit quantum tunneling effects, are critically sensitive to even the slightest impurities or defects in their constituent materials. Developing specialized superconductors and interfaces that minimize energy loss and decoherence is therefore vital for pushing the performance limits of future quantum processors and unlocking their full potential.

I am deeply interested in the development of epitaxially pure and foundry-compatible fabrication processes for quantum devices based on emerging Josephson junctions, as these efforts bridge the gap between fundamental material science and scalable quantum technology. My focus is on leveraging state-of-the-art design tools and high-precision measurement capabilities, aiming to pinpoint the underlying mechanisms that dictate coherence, fidelity, and stability in these systems. This holistic approach—spanning device fabrication, design optimization, and rigorous process control—enables the continuous refinement of device architectures and material choices, ultimately guiding the development of more robust and scalable quantum platforms.

References:

In prep.

The quantum internet is a revolutionary concept that aims to leverage the principles of quantum mechanics to create a new paradigm for communication. Quantum entanglement is harnessed for creating ultra-secure communication channels, as any attempt to eavesdrop on the transmission would disrupt the entanglement and be immediately detected. The quantum internet promises to revolutionize data security and enable new forms of communication that are currently impossible with classical technologies.

Quantum networks are the building blocks of the quantum internet, consisting of nodes connected by quantum channels that facilitate the transmission of qubits. These networks can range from small-scale, local setups connecting devices within a lab, to expansive, long-distance networks linking different cities or even countries. The potential applications of quantum networks are vast, including secure communication through Quantum Key Distribution (QKD), distributed quantum computing, and enhanced precision in sensing and metrology.

References:

Nature Communications 12 (6383)

In prep.

The parametric process enables coherent energy transfer between electromagnetic waves by capitalizing on nonlinearities, thereby enabling quantum-limited parametric amplification. Microwave parametric amplification is not only instrumental in enhancing the readout capabilities of solid-state and nanomechanical quantum devices but is also gaining significance in emerging technologies like hybrid quantum systems and axion dark matter detection. Hence, it is crucial to pursue the development of amplifiers with exceptional performance attributes, encompassing low noise, a wide bandwidth, and high saturation power. Furthermore, to facilitate the seamless integration of parametric amplifiers, it is imperative that their design and fabrication processes are both robust and straightforward.

We have engineered an innovative Josephson Parametric amplifier design using a broadband CPW-based impedance transformation technique, which amalgamates the best features of prior designs. As we move forward, my focus is on introducing novel components to enhance the amplifier's ability to handle power and to create cascaded arrangements of diverse amplifiers. This approach aims to achieve remarkable gain while maintaining the amplifier's noise at the quantum level.

References:

Physical Review Research 6, (L012035)