A layer of material just a few atoms thick inside the Josephson junction may be the key limitation preventing the next leap in quantum computing power, according to researchers at UC Santa Barbara. The university’s team joins the new $27.9 million MARQUIS institute, funded by the National Science Foundation, which brings together 24 laboratories across nine institutions to address this challenge.
“The Josephson tunnel junction is now, finally, the weakest element in a superconducting quantum processor,” says Nobel laureate Michel Devoret, a UCSB physics professor and pioneer in superconducting circuits. MARQUIS will focus on rebuilding this critical component, seeking improvements to the design after decades of progress in other areas of quantum computing.
UCSB Team Targets Josephson Junction Limitations in $27.9M Institute
The collaborative approach at UC Santa Barbara, central to the new MARQUIS institute, prioritizes interdisciplinary problem-solving, bringing together expertise rarely found in a single location. Professors Devoret, Palmstrøm and Van de Walle combine understanding of quantum device behavior, atomic-level material observation and novel material structure construction. Their combined skills are actively supported by the university.
This model, according to a UCSB representative, provides the “freedom and tools to attack” complex challenges “from fundamentally different directions.” The institute, funded by a $27.9 million National Science Foundation grant over five years, will use this approach to address limitations within Josephson junctions, the core of superconducting qubits. The focus on the Josephson junction stems from a shift in the broader field; previously dominant obstacles like electromagnetic noise and wiring losses have diminished, revealing the junction as the primary bottleneck.
This clarity allows for targeted material science investigations, providing a focused testbed for evaluating experimental junctions without interference from other device shortcomings. The team aims to move beyond incremental improvements, seeking to replicate the consistency and quality already achieved in conventional semiconductor manufacturing. Achieving this consistency is critical for scaling up quantum processors, as current algorithms demand substantial improvements in two-qubit quantum operation error correction.
The researchers estimate that Josephson tunnel junction quality must improve by a factor of ten to meet the needs of these advanced algorithms. Van de Walle’s computational work will be instrumental in understanding material behavior at the atomic scale, particularly the impact of defects. His prior research on light-emitting diodes revealed that even defects present in concentrations of less than one in a million atoms can measurably reduce efficiency, a finding with direct relevance to junction performance.
The ideal outcome, according to the team, involves fabricating single-crystal superconductors and dielectric heterostructures, eliminating disorder at interfaces and within the dielectric layers. This level of control represents a significant manufacturing challenge, but one the MARQUIS institute is uniquely positioned to address through its collaborative structure and the combined expertise of its researchers.
Nathalie Leon of Princeton has noted the reliance on the same junction technology for approximately 25 years, indicating its initial effectiveness but also signaling that innovation is now essential for continued progress. The institute’s work at UCSB builds on a long history of transitioning promising laboratory devices into reproducible technologies, an important step toward realizing the full potential of quantum computing.
The Josephson tunnel junction is now, finally, the weakest element in a superconducting quantum processor.
Michel Devoret, physics professor at UC Santa Barbara
Devoret, Palmstrøm, and Van de Walle’s Complementary Quantum Capabilities
Herbert Kroemer Distinguished Professor Chris Van de Walle’s computational work at UCSB will focus on identifying the microscopic origins of material defects that limit quantum processor performance, a challenge he previously tackled with success in light-emitting diodes. This understanding of defect behavior is critical, as identifying the microscopic origin of two-level systems has remained a great challenge in condensed-matter physics.
Palmstrøm’s group is designing fabrication processes compatible with existing semiconductor manufacturing tools, including those used for 300-millimeter silicon production, to improve the consistency of superconducting materials. “What would be ideal is to have single crystal superconductors and dielectric heterostructures so that there is no disorder in the dielectric and at the interfaces,” Palmstrøm said, acknowledging the difficulty of combining materials with differing crystal structures and chemical bonds.
The approach uses UCSB’s long history of transitioning laboratory devices into reproducible manufacturing processes, building on decades of semiconductor research. “For the Josephson junctions themselves, it would be a huge success if they could be manufactured with the same level of consistency and quality as has been achieved in semiconductor technology,” Van de Walle stated, highlighting the ambition to replicate the precision of established semiconductor fabrication.
“The useful algorithms require quantum error correction of 2-qubit quantum operations, which itself requires improvement of Josephson tunnel junction quality by a factor 10,” the team reports. This level of precision is now within reach, according to the researchers, as they have successfully mitigated other potential failure modes.
“It is only now in 2026 that we know how to mitigate all other possible failure modes, allowing us to address the problem of making the best possible junction,” they explain. Success on this front, they believe, would have an impact well beyond the goals of this specific program.
The useful algorithms require quantum error correction of 2-qubit quantum operations, which itself requires improvement of Josephson tunnel junction quality by a factor 10.
Michel Devoret, physics professor at UC Santa Barbara
Molecular Beam Epitaxy Aims for Orderly Dielectric Heterostructures
Molecular beam epitaxy will be central to fabricating single-crystal dielectric layers within Josephson junctions, a strategy aimed at minimizing energy-absorbing defects. Chris Palmstrøm’s group at UC Santa Barbara will use seven interconnected MBE systems to grow these structures layer by layer, enabling precise control over material combinations and interfaces. The approach allows for samples to move under ultrahigh vacuum, facilitating the creation of exceptionally pure single-crystal silicon as a potential replacement for aluminum oxide, the current insulating material in many junctions.
This substitution could allow for a thicker insulating layer, reducing the impact of minute thickness variations. The potential for miniaturization is significant; Palmstrøm’s group is developing a design where a fin etched from a silicon wafer hosts deposited superconducting material, potentially reducing the footprint of a transmon qubit by more than two orders of magnitude. These calculations will guide experimentation, identifying promising material combinations and explaining performance differences.
Now, with the addition of Michel Devoret’s expertise, the institute aims to establish experimental guardrails to accurately assess improvements in the junction itself, independent of other system limitations. The institute’s focus on reproducible manufacturing, using techniques from semiconductor processing, aims to transition promising laboratory results into consistently high-quality devices.
It is only now in 2026 that we know how to mitigate all other possible failure modes that we can attack the problem of making the best possible junction.
Michel Devoret, physics professor at UC Santa Barbara
Improved Junction Quality Critical for Scalable Quantum Error Correction
The collaborative approach central to MARQUIS is a deliberate strategy, mirroring a successful model at UC Santa Barbara. A key challenge lies within the atomic structure of the insulating layer in conventional junctions, typically aluminum oxide. This layer, approximately one nanometer thick, lacks the crystalline order found in ideal materials, creating microscopic configurations capable of disrupting qubit control. These two-level systems absorb energy at frequencies used to manipulate quantum bits, introducing errors.
Establishing reliable metrics for improvement is paramount, and Michel Devoret’s role within the institute is specifically geared toward that goal. Devoret will help establish experimental “guardrails” to differentiate genuine improvements in junction quality from confounding factors elsewhere in the quantum circuit or measurement systems. The institute’s ultimate goal extends beyond creating a single high-performing junction. It aims to achieve consistent, reproducible fabrication.
Nathalie de Leon of Princeton, who directs the institute, has observed that the foundational junction technology has remained largely unchanged for approximately 25 years. While adequate for current experimental systems, future processors with significantly larger qubit counts will necessitate tighter control over device characteristics.
This transition from promising laboratory prototypes to reliably manufactured components builds on a long-standing strength at UCSB, where fundamental understanding of materials and fabrication processes is prioritized. The potential benefits of improved junctions are substantial, potentially reducing the number of physical qubits needed for error-corrected logical qubits and simplifying the associated control and refrigeration systems.
Professors Devoret, Palmstrøm and Van de Walle bring together three capabilities that rarely exist in one place: understanding how quantum devices behave, seeing what is happening at the atomic level and being able to build entirely new material structures.
Umesh Mishra, dean of the Robert Mehrabian College of Engineering
See today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals.
