Quantum computing is a very interesting field, due to the ability of quantum computers to solve many problems much faster than a classical computer. Superconducting qubits are electronic circuits composed of superconducting capacitors, inductors and Josephson junctions, which can implement a physical qubit. However they need improvements in their coherence time to create a viable quantum computing. In this work I study the effect on decoherence caused by two level systems in the native oxide that forms on the surface of qubit. To this end I fabricate 3D transmon qubits using materials grown my molecular beam epitaxy, to which a variety of different surface treatments had been applied. I began by fabricating qubits from niobium/aluminum oxide/niobium trilayers. To this end I developed a self-aligned process for fabricating sub-micron Josephson junctions. This process presented many challenges. During the development of this process, it became clear that niobium was an inferior material for fabricating qubits compared to aluminum. I then switched to making qubits from Aluminum. I began by studying the growth of aluminum on sapphire, and was able to achieve aluminum films with an RMS roughness of 0.2 nm by growing on c-plane sapphire that had been annealed in oxygen at 1100 °C and dipped in BOE. Next I fabricated 3D transmon qubits by adapting the standard shadowmask process for use with MBE. I fabricated qubits with a long in situ oxidation to fully passivate the surface before exposure to air. I also passivated the surface by means of growing co deposited aluminum oxide, and by grown aluminum nitride using a nitrogen plasma source. The coherence times of these qubits were compared to those a control sample that had been exposed to air immediately after growth. Overall it doesn’t appear that cladding the surface changes the coherence time much, however it is difficult to form conclusions with this small sample size. The coated samples appeared to have slightly longer coherence times than the untreated samples, with diffusive oxidation and coating with aluminum oxide providing the largest increase in T1 and diffusive oxidation providing the largest increase in T2.
Quantum computing is a very interesting field, due to the ability of quantum computers to solve many problems much faster than a classical computer. Superconducting qubits are electronic circuits composed of superconducting capacitors, inductors and Josephson junctions, which can implement a physical qubit. However they need improvements in their coherence time to create a viable quantum computing. In this work I study the effect on decoherence caused by two level systems in the native oxide that forms on the surface of qubit. To this end I fabricate 3D transmon qubits using materials grown my molecular beam epitaxy, to which a variety of different surface treatments had been applied. I began by fabricating qubits from niobium/aluminum oxide/niobium trilayers. To this end I developed a self-aligned process for fabricating sub-micron Josephson junctions. This process presented many challenges. During the development of this process, it became clear that niobium was an inferior material for fabricating qubits compared to aluminum. I then switched to making qubits from Aluminum. I began by studying the growth of aluminum on sapphire, and was able to achieve aluminum films with an RMS roughness of 0.2 nm by growing on c-plane sapphire that had been annealed in oxygen at 1100 °C and dipped in BOE. Next I fabricated 3D transmon qubits by adapting the standard shadowmask process for use with MBE. I fabricated qubits with a long in situ oxidation to fully passivate the surface before exposure to air. I also passivated the surface by means of growing co deposited aluminum oxide, and by grown aluminum nitride using a nitrogen plasma source. The coherence times of these qubits were compared to those a control sample that had been exposed to air immediately after growth. Overall it doesn’t appear that cladding the surface changes the coherence time much, however it is difficult to form conclusions with this small sample size. The coated samples appeared to have slightly longer coherence times than the untreated samples, with diffusive oxidation and coating with aluminum oxide providing the largest increase in T1 and diffusive oxidation providing the largest increase in T2.
Superconducting qubits in today's quantum processing units are typically fabricated with angle-evaporated aluminum--aluminum-oxide--aluminum Josephson junctions. However, there is an urgent need to overcome the limited reproducibility of this approach when scaling up the number of qubits and junctions. Fabrication methods based on subtractive patterning of superconductor--insulator--superconductor trilayers, used for more classical large-scale Josephson junction circuits, could provide the solution but they in turn often suffer from lossy dielectrics incompatible with high qubit coherence. In this work, we utilize native aluminum oxide as a sidewall passivation layer for junctions based on aluminum--aluminum-oxide--niobium trilayers, and use such junctions in qubits. We design the fabrication process such that the few-nanometer-thin native oxide is not exposed to oxide removal steps that could increase its defect density or hinder its ability to prevent shorting between the leads of the junction. With these junctions, we design and fabricate transmon-like qubits and measure time-averaged coherence times up to 30 μs at a qubit frequency of 5 GHz, corresponding to a qubit quality factor of one million. Our process uses subtractive patterning and optical lithography on wafer scale, enabling high throughput in patterning. This approach provides a scalable path toward fabrication of superconducting qubits on industry-standard platforms.
Abstract We present a transmon qubit fabrication technique that yields systematic improvements in T 1 relaxation times. We encapsulate the surface of niobium and prevent the formation of its lossy surface oxide. By maintaining the same superconducting metal and only varying the surface, this comparative investigation examining different capping materials, such as tantalum, aluminum, titanium nitride, and gold, as well as substrates across different qubit foundries demonstrates the detrimental impact that niobium oxides have on coherence times of superconducting qubits, compared to native oxides of tantalum, aluminum or titanium nitride. Our surface-encapsulated niobium qubit devices exhibit T 1 relaxation times 2–5 times longer than baseline qubit devices with native niobium oxides. When capping niobium with tantalum, we obtain median qubit lifetimes above 300 μs, with maximum values up to 600 μs. Our comparative structural and chemical analysis provides insight into why amorphous niobium oxides may induce higher losses compared to other amorphous oxides.
Improving qubit lifetime is essential for fault-tolerant quantum computing. It is widely acknowledged that intricate processing techniques introduce defects at the interfaces and surfaces of superconducting quantum circuits, which can create sources of decoherence. Therefore, a precise understanding of how a material's structure contributes to decoherence is critical for enhancing the performance of superconducting qubits. This talk highlights our recent advancements in understanding surface oxides in resonators and ultrathin aluminum oxide barriers in Josephson junctions (JJs) using transmission electron microscopy and spectroscopy techniques [1-4]. The Superconducting Quantum Materials & Systems Center (SQMS) demonstrated that encapsulating niobium surfaces with a ∼10nm tantalum layer results in 2-5x longer T1 than bare niobium [2]. Our collaborative investigation shows that this improvement is attributed to tantalum's less lossy surface oxide, which contains fewer sub-oxides than niobium oxide. Moreover, the amorphous Ta2O5 shows a bonding nature closer to crystalline form than amorphous Nb2O5, potentially reducing hydrogen diffusion toward the oxide/metal interface [1, 2]. Moreover, we will briefly discuss our preliminary study of the thin film structure under cryogenic conditions down to 20K, conducted using a liquid helium TEM holder in a Thermo Cs-corrected Titan. We also explored the effect of aluminum deposition rates on aluminum oxide microstructure and qubit lifetime. Although increasing the deposition rate from 0.5 Å/s to 5 Å/s significantly improved oxide layer roughness and thickness variation, the qubits’ coherence remained unchanged. This is likely due to Cooper pairs only tunneling through the thinnest parts of the barrier, making them less affected by the grain boundaries in the thin aluminum films. Additionally, we uncovered a stress-induced grain boundary sliding mechanism, which contributes to short-circuit failures in JJs [4]. Our discovery provides vital insight into oxide microstructure and superconducting qubits’ functionality [5].
Quantum computation is a fascinating field that combines novel physics with improvements in computation times and has been rapidly growing in the past few years. Using superconductors to form the qubits has the potential for large-scale computing, if the decoherence inherent in these devices can be understood and reduced. Two-level fluctuators due to defects in the materials are thought to cause changes to the Josephson critical current or the flux through the superconducting loop of a flux qubit, which leads to decoherence in the qubit. Alternatively, defects in the crystal lattice give rise to electron localization which in turn traps spins with random orientations at the substrate/metal interface, again producing decoherence. My work studied the proposed noise mechanisms by using Molecular Beam Epitaxy to fabricate single-crystal niobium-based Josephson junctions. Using RHEED, AFM, and TEM/STEM, I studied the epitaxy of the niobium film at the substrate interface to reduce noise due to crystal defects. I then measured flux noise in these films; the noise in the epitaxial films is lower than the comparative polycrystalline films. Further measurements using ex-situ Josephson junctions and epitaxial niobium loops resulted in the lowest reported flux noise measurements to date. Additionally, EELS measurements made in the course of the STEM analysis of the crystal structure reveal oxygen depletion from the substrate at elevated growth temperatures. This depletion leads to oxygen vacancies in the aluminum oxide substrate, which can in turn lead to charge traps at the substrate/metal interface and hence to decoherence in the qubit. Critical current noise was investigated by changing the oxidation dose of the alumina barrier in more than 50 separate Josephson junctions. Films which were grown with lower oxidation doses have a non-ideal critical current temperature dependence as well as higher critical current noise values, suggesting that insufficient oxygen is being incorporated into these junctions. Finally, I studied the effect of surface cleaning after fabrication for resonators and transmons. Microscopy shows that an oxygen ash and BOE dip together removes all photoresist residue from the chip. Internal-Q measurements do not show a marked improvement due to this cleaning, but the coherence times do improve upon the cleaning step.
While significant strides have been made to increase the coherence time of superconducting qubits, further advancements are essential for realizing scalable quantum computing. Decoherence is often a result of loss and noise stemming from two-level systems and excess quasiparticles, arising due to material defects, fabrication processes, and ambient exposure, particularly at surfaces and interfaces. Our recent efforts to mitigate these decoherence mechanisms have employed a variety of strategies, including low-loss surface encapsulation materials, advanced substrate preparation techniques, modifications to metal film growth, and the development of novel fabrication processes. The structural and chemical properties of materials, surfaces, and interfaces are studied using scanning probe microscopy, electron microscopy, photoelectron spectroscopy, mass spectrometry, and X-ray diffraction, which is correlated to device performance metrics, including superconducting resonator internal quality factor and qubit T1 time. This information is used to identify and understand material sources of loss and their origins in the device fabrication process. Through multi-institution efforts within SQMS we have identified the loss mechanism of interstitial hydrogen in niobium-based devices and shown how standard fabrication processes introduce these hydrides, developing strategies to mitigate their formation.1 Furthermore, we have characterized the metal-substrate interface, including the loss of niobium-silicides formed at that interface, and developed silicon surface treatments that reduce atomic scale roughness and oxygen content at the metal-substrate and Josephson junction interfaces.2-4 By developing the connection between materials properties and the overall performance of superconducting quantum circuitry, we can develop fabrication strategies to mitigate material losses, thus supporting the ongoing efforts to enhance coherence time in superconducting quantum devices. 1. Torres-Castanedo, C. G.*, Goronzy, D. P.*, et al., Adv. Funct. Mater., 2401365 (2024) 2. Lu, X., et al., Phys. Rev. Materials 6, 064402 (2022) 3. Berti, G., Appl. Phys. Lett. 122, 192605 (2023) 4. Kopas, C. J., Goronzy, D. P., et al., arXiv:2408.02863 (2024)
Niobium is commonly used for superconducting quantum systems as readout resonators, capacitors, and interconnects. The coherence time of the superconducting qubits is mainly limited by microwave dissipation attributed to two-level system defects at interfaces, such as the Nb/Si and Nb/air interface. One way to improve the Nb/air interface quality is by thermal annealing, as shown by extensive studies in 3D superconducting radio frequency (SRF) cavities. However, it is unclear how the microstructure and chemistry of the interface structures change during heat treatment. To address this knowledge gap, we comprehensively characterized Nb films deposited on Si wafers by physical vapor deposition, including (1) an Nb film from a transmon and (2) an Nb film without any patterning step, using an aberration-corrected transmission electron microscope. Both Nb films exhibit columnar growth with strong [110] textures. There is a double layer between the Nb film and Si substrate, which are amorphous niobium silicides with different Nb and Si concentrations. After in-situ heating of the heterostructure at 360°C inside the microscope, the composition of the double layers at the Nb-Si interface remains almost the same despite different thickness changes. The initial amorphous niobium oxide layer on Nb surface decomposes into face-centered cubic Nb nanograins in the amorphous Nb-O matrix upon heating.