Resolving the positions of defects in superconducting quantum bits

Abstract

Solid-state quantum coherent devices are quickly progressing. Superconducting circuits, for instance, have already been used to demonstrate prototype quantum processors comprising a few tens of quantum bits. This development also revealed that a major part of decoherence and energy loss in such devices originates from a bath of parasitic material defects. However, neither the microscopic structure of defects nor the mechanisms by which they emerge during sample fabrication are understood. Here, we present a technique to obtain information on locations of defects relative to the thin film edge of the qubit circuit. Resonance frequencies of defects are tuned by exposing the qubit sample to electric fields generated by electrodes surrounding the chip. By determining the defect's coupling strength to each electrode and comparing it to a simulation of the field distribution, we obtain the probability at which location and at which interface the defect resides. This method is applicable to already existing samples of various qubit types, without further on-chip design changes. It provides a valuable tool for improving the material quality and nano-fabrication procedures towards more coherent quantum circuits.Material defects have a manifold of microscopic origins such as impurities in solids or adsorbates hosted on surfaces [1]. Their detrimental role was identified already in first experiments with superconducting quantum bits (qubits) [2]. Strong interaction with a long-known defect type, charged two-level systems (TLS) [3,4], residing in the tunnel barrier of a qubit's Josephson junction gives rise to avoided level crossings and resonant energy absorption [5]. This form of dielectric loss could be mitigated by reducing the amount of lossy dielectrics, e.g. by incorporating smaller Josephson junctions [6] and by avoiding insulating layers. Another strategy is to enlarge the footprint of device capacitors in order to dilute the electric field induced by the qubit, which excites defects by coupling to their electric dipole moments [7,8]. As a consequence of significantly enhanced coherence times, qubits became sensitive also to weakly coupling defects residing on the surfaces and interfaces of circuit electrodes [9]. Since these are limiting the performance of state-of-the-art circuits [10][11][12], further progress towards scaled-up quantum processors requires strong efforts to prevent the appearance of defects, e.g. by using better materials, improved fabrication procedures [13][14][15][16], and surface treatment to avoid contamination and parasitic adsorbates [17,18]. This endeavor needs to be guided by careful analysis of defect properties such as densities, electric dipole moments, and positions, in order to identify and improve the manufacturing steps that reduce defect formation, and to analyze the microscopic structure of defects.In this Letter, we present a method to extract information on the spatial positions of defects at the profile of the film edge in a qubit circuit. For doing this, we exploit the tunabi...

Two-level systems in superconducting quantum devices due to trapped quasiparticles

“…It is becoming increasingly evident that the most coherence-limiting two-level systems (TLS) reside outside the qubit junctions on the surface of metals and dielectrics surrounding them (1)(2)(3)(4)(5) and their slowly fluctuating dynamics poses a serious challenge for quantum computation (1,(5)(6)(7)(8). A wide range of techniques has been developed to study such TLS by developing qubits and resonators into probes of a wide range of material properties (2,5,(9)(10)(11)(12)(13)(14)(15).…”

In-situ bandaged Josephson junctions for superconducting quantum processors

“…and contamination at electrode interfaces are major limiting factors for qubit coherence [4][5][6][7].…”

Identification of different types of high-frequency defects in superconducting qubits

“…Extensive studies in recent decades have identified major noise sources in superconducting qubits, including * leonid.abdurakhimov.nz@hco.ntt.co.jp parasitic two-level-system (TLS) defects [13][14][15][16][17][18][19][20][21][22][23][24][25][26], quasiparticle noise [27][28][29][30][31][32][33][34][35][36][37], and, for flux-tunable qubits, a 1/f magnetic flux noise [38,39]. In this paper, we focus on TLS defects: defects of different microscopic nature that can be modeled as two-level quantum systems, such as tunneling atoms, dangling electronic bonds, impurity atoms, and trapped charge states [21].…”

“…Normally, it is assumed that the qubit-defect interaction is caused by an electric-field coupling between an electric dipole associated with a charge TLS defect and a microwave electric field generated across a Josephson junction [14]. The standard techniques to study high-frequency TLS defects in superconducting qubits are based on matching the frequencies of qubit and defect transitions either by adjusting the qubit frequency with an applied magnetic flux [13,14,16,45,46] or changing the defect frequency with an external electrical field or an applied mechanical strain [17,[22][23][24]. Parameters of a TLS defect are then extracted from the splitting of an avoided crossing observed in the qubit spectrum [13,14], the qubit response to a Rabi drive [45], or measurements of energyrelaxation times [22][23][24].…”

“…The standard techniques to study high-frequency TLS defects in superconducting qubits are based on matching the frequencies of qubit and defect transitions either by adjusting the qubit frequency with an applied magnetic flux [13,14,16,45,46] or changing the defect frequency with an external electrical field or an applied mechanical strain [17,[22][23][24]. Parameters of a TLS defect are then extracted from the splitting of an avoided crossing observed in the qubit spectrum [13,14], the qubit response to a Rabi drive [45], or measurements of energyrelaxation times [22][23][24]. However, the standard methods of TLS defect spectroscopy based on the qubit frequency tuning are not suitable for fixed-frequency qubits, such as fixed-frequency transmons.…”