Atomic tunneling Systems and fluctuating Spins interacting with superconducting Qubits

While atomic tunneling systems emerging from material defects have been identified as a major source of decoherence in superconducting quantum circuits, their microscopic origins and locations in a given device remain unknown. Here, we demonstrate a new technique to tune the resonance frequencies of defects in a transmon qubit by DC-electric fields which are generated by electrodes surrounding the sample chip. In addition, defects are tuned by mechanical strain generated by a piezo actuator. A comparison of the observed response to electric field and strain allows us to distinguish defects in the tunnel barriers of qubit junctions from those residing in surface oxides or emerging from adsorbates. Our statistical analysis indicates that these two defect classes contribute approximately equally to qubit decoherence. Moreover, we present a method to obtain information about the locations of individual defects on the sample surface by comparing their coupling strengths to different DC-electrodes. These techniques can be applied to any ready-made qubit sample and provide valuable tools for the improvement of qubit fabrication procedures by helping one to identify the critical interfaces where TLS participation dominates.

Two level systems (TLSs) in amorphous materials and at material interfaces have long plagued planar superconducting circuits. These states interfere with qubit lifetimes and often set the limit on resonator quality factors in the single photon regime. We designed superconducting circuits that enable characterization of these states for planar devices. In one such example we designed a bias bridge resonator circuit that is sensitive to TLSs on surfaces. We will present data on dipole moment extraction and deviations from standard model behavior. In another experiment we measured TLSs for amorphous silicon grown using electron evaporation. We then compared the results to acoustic properties measured using the same growth methods. Finally, we will discuss other experiments and methods that we are pursuing at Lawrence Livermore to probe and model the noise and loss mechanisms in superconducting circuits.

The nature of the low energy excitations (LEEs) that are responsible for the low temperature thermal and mechanical behavior of glass is a long-standing open question. The standard tunneling model proposes that the LEEs are atomic-scale tunneling two level systems (TLSs) and successfully explains much of the low temperature behavior of glass. In recent ground-breaking experiments, individual TLSs were probed in the amorphous tunnel junction of superconducting qubits. The TLSs that were directly measured in these experiments were electrically coupled to the qubit. In contrast, we are developing a system for measurements of individual TLSs that need not respond to electric fields. Such measurements will allow direct comparison to the large body of mechanical measurements of bulk glass. The glass samples containing the TLSs will be nanomechanical resonators (NEMS), and the strain coupling between the mechanical mode and the TLS will be used to control the quantum state of the TLS. In order to resolve individual TLSs, our NEMSs, with a fundamental resonances around 100 MHz, must be cooled to their ground state, T << 5 mK. We will discuss the characteristics of our NEMS as well as the development of a unique ultra-low temperature optomechanical measurement system.

Atomic two-level system (TLSs) are known to cause performance limiting decoherence in qubits, and phase noise in astronomy photon detectors. We have developed different methods of coherently inducing excitations in these TLSs. These studies test the dynamics of these tunneling systems, and also the two-level nature in a broad TLS distribution. This is complementary to the coherent approach of studying TLSs as quasi-static levels for interactions with qubits, resonators, etc. At first, I will describe a universal property which can be tested in any amorphous film which is related to driving TLSs with a combination of swept and microwave fields. We confirm that this relationship holds in most of our dielectric films. I will then go beyond this to describe other phenomena which this population control allows. For example, in a related setup we are able to obtain lasing from a net inversion of TLSs, which on the face of it seems impossible since the distribution of TLSs is random in energy and angular orientation. Here rapid adiabatic passage allows a net inversion and a microwave resonator accepts the stimulated emission from the excited TLSs. This allows coherence caused by TLSs, whereas TLSs are normally infamous as decohering defects! In newer experiments, we coherently excite TLSs over a broadband, many GHz near a microwave resonator, which induces strong dispersion from the “inverted dispersion” of excited TLSs. We also study defects in new regimes which appear to reveal phenomena which is clearly distinguishable from the standard model of TLSs.

New techniques to identify the location of decoherence-inducing material defects known as Two-Level-Tunneling systems (TLS) in superconducting qubits are demonstrated. We expose a transmon qubit circuit to a DC-electric field generated by electrodes surrounding the sample chip, and study the TLS response by monitoring their resonance frequencies using qubit swap spectroscopy. We find that about $50\%$ of all detectable TLS do not couple to the applied E-field, as it is expected from TLS hosted in the tunnel junction barrier, but unlikely for TLS residing in surface oxides or at substrate interfaces. In contrast, all TLS respond to the mechanical strain generated by a piezo actuator. This indicates that surface TLS contribute about equally to qubit decoherence as those in qubit junctions. Moreover, by comparing measured and simulated coupling strengths to each DC-electrode, we obtain information about the possible locations and hosting interfaces of the observed surface TLS. This kind of analysis directly indicates which circuit interfaces must be improved in order to enhance qubit coherence.\\ \textbf{A. Bilmes}, G. Weiss, A. V. Ustinov, R. Barends, J. Kelly, A. Megrantand, and J. Lisenfeld

Superconducting qubits are an attractive platform for quantum computing since they have demonstrated high-fidelity quantum gates and extensibility to modest system sizes. Nonetheless, an outstanding challenge is stabilizing their energy-relaxation times, which can fluctuate unpredictably in frequency and time. Here, we use qubits as spectral and temporal probes of individual two-level-system defects to provide direct evidence that they are responsible for the largest fluctuations. This research lays the foundation for stabilizing qubit performance through calibration, design, and fabrication.

An on-chip electron spin resonance technique is applied to reveal the nature and origin of surface spins on Al2O3. We measure a spin density of 2.2 × 1017 spins/^m2, attributed to physisorbed atomic hydrogen and S = 1/2 electron spin states on the surface. This is direct evidence for the nature of spins responsible for flux noise in quantum circuits, which has been an issue of interest for several decades. Our findings open up a new approach to the identification and controlled reduction of paramagnetic sources of noise and decoherence in superconducting quantum devices.

Material defects in the form of two-level systems (TLS) with energy splittings in the microwave range couples in large numbers to superconducting resonators through their electric dipoles, causing both excess loss and frequency fluctuations. Here we explore the noise in superconducting resonators by simultaneously measuring noise and the electron spin resonance (ESR) spectrum of the resonator [1]. By desorbing some of the surface spins, these experiments reveal a possible link between the number of paramagnetic defects present and the magnitude of dielectric (frequency) noise in the devices. Hence, ESR using superconducting resonators turns out to be a powerful tool to reveal not only possible sources of flux noise, but also the identity of defects coupling to quantum devices through their electric dipole moments. We also discuss the ongoing work in developing a quantum limited near-field scanning microwave microscope (NSMM) [2] based on a micro-machined superconducting resonator as the scanning probe tip, with the ultimate ambition to image a single TLS defect on a substrate. [1] S.E. de Graaf et al. Nature Communications 9:1143 (2018); S.E. de Graaf et al., APL 113, 142601 (2018) [2] S.E. de Graaf et al. Scientific Reports 5, 17176 (2015)

We report on long-term measurements of a highly coherent, non-tunable transmon qubit, revealing low frequency burst noise in coherence and transition frequency. We present a simultaneous measurement of the qubit’s relaxation and dephasing rates as well as resonance frequency fluctuations, and analyze their correlations. These yield information about the microscopic origin of the intrinsic decoherence mechanisms in Josephson qubits and their fluctuation dynamics. From a spectral noise analysis we obtain further evidence for our presented model of a small number of dominant fluctuators.

The superconducting Josephson junction qubit is one of the leading candidates for making a qubit. A major obstacle to the realization of quantum computers with Josephson junction qubits is noise and decoherence of the wavefunction. One of the major sources of noise is called flux noise and is due to fluctuating magnetic spins on the surface of metals. We will describe our work on the microscopic sources of this flux noise; namely oxygen molecules adsorbed on the surface as well as atomic hydrogen.

Pure phase decoherence (dephasing), where the quantum state is decohered without any energy transfer to or from the system (that is, without transitions beween the eigenstates) is quite common i two level systems, and is well understood. In a two level system, the only way to achieve this is when the noise term (or interaction term for a quantum environment) in the Hamiltonian is proportional to the system Hamiltonian. The situation is richer for systems with more than two levels, because of the existence of Hermitian operators which commute without being proportional. A special case of flux noise through a tight-binding ring of atoms was analyzed by Zhu et al., Phys. Rev. A 81, 062127 (2010). We have studied the same model for small numbers of atoms, and find that we can give a clear geometrical interpretation of the observed dynamics in terms of the generalized Bloch vector. In particular, we can explain the persistent oscillations of the occupation numbers of the sites as a consequence of the coupling term in the Hamiltonian commuting with the system Hamiltonian.

The last decades have seen significant advances in the coherence times of superconducting qubits. This was mainly made possible by reducing the charge dispersion of transmon qubits, better thermalization and filtering of the readout and control circuitry as well as improvements in Josephson junction fabrication. Lately, efforts are being made in order to investigate and understand the limitations of coherence due to material interfaces and two level fluctuators [1-3]. Here we present our work towards understanding decoherence mechanisms in transmon qubits via participation ratio engineering, surface treatments and sample packaging. We show qubit packaging under UHV or a controlled atmosphere, study possible surface treatments and discuss coherence data from initial devices. Additionally, we present preliminary steps towards reducing the effects of two-level fluctuators by investigating different device geometries. [1] J. M. Gambetta, et al, IEEE Transactions on Applied Superconductivity, 27, 1700205, (2016). [2] S. De Graaf et al., Nat. Communications 9, 1143, (2017) [3] C. Müller et al., arXiv:1705.01108 (2017)

We study the temporal stability of relaxation and dephasing in superconducting qubits. By collecting statistics during measurements spanning multiple days, we reveal large fluctuations of qubit lifetimes and find that the cause of T1 fluctuations is interacting parasitic two-level-systems (TLS). Our statistical analysis also provides useful information about the dynamics of TLS which could help identify possible microscopic sources of TLS, that still remain elusive. Moreover, interacting TLS also causes capacitance fluctuations, ultimately leading to frequency noise and dephasing of the qubit state. These discoveries are important for manufacturing stable superconducting circuits suitable as a scalable quantum computing platform. Here, knowledge of the timescales for drift and fluctuations are crucial to optimise calibration intervals and therefore reduce unnecessary downtime.

This work is concerned with Al/Al-oxide(AlOx)/Al-layer systems which are important for Josephson-junction-based superconducting devices such as quantum bits. The device performance is limited by noise, which has been to a large degree assigned to the presence and properties of two-level tunneling systems in the amorphous AlOx-tunnel barrier. This has motivated our study on the optimization of the structural properties of Al/AlOx/Al-layer systems and the nature of the atomic tunneling systems contained in these layers. The structures were fabricated in a high-vacuum electron-beam deposition system. Analysis was performed by transmission electron microscopy combined with electron energy loss spectroscopy for composition determination of the AlOx-tunnel barrier. We show that the fabrication conditions have a strong impact on the nanostructural and nanochemical properties of the Al/AlOx/Al-layer system. Pretreatment of Si(111) substrates and deposition conditions were optimized to obtain quasi-epitaxial Al growth resulting in a low grain boundary density of the lower Al electrode as a prerequisite for AlOx-tunnel barriers with homogeneous thickness. Depending on the oxidation conditions, the composition of the tunnel barrier varies between AlO0.9 and AlO1.3. Low-temperature dielectric measurements show that the properties of two-level tunneling systems depend on the nanostructure and composition of the AlOx-tunnel barrier.

High-quality superconducting junctions can be obtained using proximity induced superconductivity in single walled carbon nanotubes. Combined with superconducting leads having a gap on the order of 1 meV, ample supercurrent can be reached for applications involving qubits and optomechanics. We study suspended, 300-nm-long single-walled carbon nanotubes (SWCNT) contacted using MoRe leads. Good contact transparency of the superconductor-nanotube interface allows for strong proximity-induced superconductivity in our SWCNT devices. The magnitude of the switching supercurrent ranges up to 50 nA and can be tuned periodically by gate-induced charge. The gate charge modulates the retrapping current even more strongly, and its magnitude becomes vanishingly small far away from the charge degeneracy point. One of our goals has been to understand the interplay between mechanical and superconducting degrees of freedom, in particular under rf excitation that induces resonant mechanical motion. Under rf irradiation, our SWCNT devices display clear Shapiro steps, the shape of which depend on the rf frequency and power. Under certain conditions the observed steps become hysteretic, indicating small dissipation on the Josephson phase dynamics at the rf frequency. In these SWCNT devices we find mechanical resonances around 1.5 GHz, while the Q factors amount up to 15000 near the charge degeneracy point. Mechanical modes can be observed also in the superconducting regime either by mixing by the current-phase relation or by inducing Shapiro steps resonantly with the mechanical mode, which both reflect the interplay between the Josephson dynamics and the mechanical degrees of freedom. In addition, we find bifurcation of Shapiro steps around the mechanical resonance frequency, both at fundamental and subharmonic resonance drives. Application possibilities for qubits and optomechanics will be discussed in the light of the obtained results

The efficiency of the future devices for quantum information processing will be limited mostly by the finite decoherence rates of the individual qubits and quantum gates. Recently, substantial progress was achieved in enhancing the time within which a solid-state qubit demonstrates coherent dynamics. This progress is based mostly on a successful isolation of the qubits from external decoherence sources obtained by clever engineering. Under these conditions, the material-inherent sources of noise start to play a crucial role. In most cases, quantum devices are affected by noise decreasing with frequency f approximately as 1/f. According to the present point of view, such noise is due to material- and device-specific microscopic degrees of freedom interacting with quantum variables of the nanodevice. The simplest picture is that the environment that destroys the phase coherence of the device can be thought of as a system of two-state fluctuators, which experience random hops between their states. If the hopping times are distributed in a exponentially broad domain, the resulting fluctuations have a spectrum close to 1/f in a large frequency range. I will review the current state of the theory of decoherence due to degrees of freedom producing 1/f noise. I will discuss basic mechanisms of such noises in various nanodevices and then review several models describing the interaction of the noise sources with quantum devices. The main focus of the review is to analyze how the 1/f noise destroys their coherent operation. I will start from individual qubits concentrating mostly on the devices based on superconductor circuits, and then discuss some special issues related to more complicated architectures. Finally, several strategies for minimizing the noise-induced decoherence will be considered. [1] J. Bergli, Y. M. Galperin, and B. L. Altshuler, “Decoherence in qubits due to low-frequency noise,” New J. Phys. 11, 025002 (2009). [2] E. Paladino, Y. M. Galperin, G. Falci and B. L. Altshuler, “1/f noise: implications for solid-state quan-tum information”, Rev. Mod. Phys. 86, 361 (2014).

We measure the high kinetic inductance material WSi_x, which is considered an amorphous superconducting material, patterned into GHz frequency resonators and nano-wires. Such devices have high-kinetic inductance and we thus expect to find different susceptibility to surface defects and fluctuators. We also measure the nonlinear response of a multimode device, and determine the frequency mixing coefficients.

Dielectric measurements of the dynamics of atomic tunneling systems in thin-film AlOx S. Meißner, A. Seiler, G. Weiss Karlsruher Institut für Technologie, Karlsruhe, Germany Josephson junctions made from Al/AlOx/Al film sandwiches are the central elements of superconducting quantum circuits. As all disordered solids, the non-crystalline AlOx contains atomic two-level tunneling systems (TS) which play an inglorious role in dephasing the quantum states of the circuit under consideration. Interaction is mainly caused by the electric dipole moment of the tunneling group of atoms coupling strongly to electric fields of the circuit. Here, we present a broadband study ranging from kHz to GHz frequencies of the dynamics of TS contained in thin-film disordered AlOx. The TS ́ density of states is probed at kHz frequencies by capacitance measurements as well as in the GHz range by tracking the resonance frequencies of superconducting microstrip resonators with embedded plate capacitors. Due to the homogeneous electric field concentrated in the dielectric AlOx the influence of native oxide on top of the Al structures is negligible. The large bandwidth of the excitation frequencies allows us allows us to verify general predictions of the standard tunneling model. Our measured data, however, is more consistent with an increasing density of TS with increasing energy. Moreover, the low frequency measurements surprisingly show an influence of a magnetic field applied to the plate capacitors. The influence is attributed to the interaction between TS and quasi particles in the superconducting state of the Al electrodes and conduction electrons in the normal state, respectively.

Recently, individual control and addressing of TLS have allowed detailed studies of the temperature and strain dependence of TLS decoherence. The interaction of TLS with phonons within the material is typically considered an important decoherence mechanism which has been studied in depth for ensembles of TLS within the framework of the standard tunnelling model (STM). More recently the interplay of two classes of TLS has been proposed as an argument for the near universal behaviour observed in amorphous materials. These two classes are distinguished by the extent to which they couple to the phonon field, one relatively weakly due to symmetry, the other more strongly. An alternative model considers the possibility that single electrons which are ‘dressed’ by the phonon field result in an effective candidate for TLS. Given the importance of understanding the interaction between TLS and phonons, raises the question, what are limits of what we can determine about TLS simply by measuring the phonon response? To answer this question requires careful consideration of the input parameters to the STM, its microscopic justification and the influence of geometry on the phonon response.

To fabricate superconducting qubits of consistent quality, it is important to be able to reproducibly manufacture junctions with identical resistances (and critical currents). Experimental investigations of Al/AlOx/Al Josephson junctions have found a Gaussian distribution of barrier thicknesses exists within individual devices [1]. The thickness variation causes subsequent variations in current density, where the thinnest regions dominate transport. Other aspects of the microscopic structure of the barrier have been considered as candidates for two-level systems (TLS) [2]. It is therefore important to understand the atomic structure of the amorphous barrier oxides, their interface with the aluminium electrodes, and the connection between the atomic structure and device performance. The aluminium-oxide is often formed using a low-pressure oxidation of the aluminium surface. In some simulations heightened pressures and temperatures are used to accelerate the oxidation dynamics [3]. In contrast we consider simulating the oxidation directly by adding individual oxygen atoms (or molecules) to the aluminium surface in an iterative procedure. This allows us to capture the low pressures used in fabrication in our model. With this approach we are able to reproduce minutes of real time oxidation by only simulating the interactions between the new oxygen atom and the surface while assuming that the system evolves minimally between deposition events. The electronic properties of the device are strongly dependent on the morphology of the barrier, both at the interfaces of the superconducting leads and within the metal oxide itself. In spite of this, the transport of charge through these devices is usually modelled with a simplified one-dimensional potential which “averages out” the effects of material defects inside the junction [4]. We calculate the I–V response and normal state resistances of atomistic junction models with the non-equilibrium Green’s functions formalism (NEGF). This calculation incorporates the full detail of the three-dimensional structure in the barrier. The effect of varying the thickness, stoichiometry and oxide density on the resistance of the tunnelling barrier is also investigated. Our electronic transport model reproduces the expected exponential dependence on barrier thickness and predicts an exponential dependence the oxide density. The end-to-end modelling approach we apply provides insight into how the junction structure at a microscopic level relates to the electronic properties of the device. -- [1] L. J. Zeng et al., “Direct observation of the thickness distribution of ultra thin AlOx barriers in Al/AlOx/Al Josephson junctions”, Journal of Physics D: Applied Physics, 48(39), pp. 395308, 2015. [2] Müller, C., Cole, J. H., & Lisenfeld, J. (2017). “Towards understanding two-level-systems in amorphous solids - Insights from quantum devices”, pp. 1–30. arXiV:1705.01108 [3] T. Campbell et al., “Oxidation of aluminium nanoclusters”, Physical Review B, 71(20), pp. 205413 (2005). [4] W. F. Brinkman, R. C. Dynes, and J. M. Rowell, “Tunneling conductance of asymmetrical barriers”, Journal of Applied Physics, 41(5), pp. 1915, 1970.

Here we treat the spectral diffusion of two level system (TLS) energies explicitly using the generalized master equation formalism to describe the non-linear absorption. The proposed theory predicts that the linear absorption regime holds while a TLS Rabi frequency is smaller than their phase decoherence rate. At higher external fields, a nonlinear absorption regime is found with the loss tangent inversely proportional to the intensity of the field while at highest intensities the earlier predicted "noninteracting" behavior is restored.

In this talk I will discuss a theoretical and experimental study of the dielectric loss of superconducting resonators in the presence of a periodic bias field. This field slowly changes the energy bias of tunneling two-level systems (TLSs), and sweeps them through resonance with the resonator. The dynamics of each transition is of the Landau-Zener type. To obtain the dielectric loss, we calculate the full-counting statistics of the number of photons absorbed by a single TLS. At low and intermediate sweep rates, the theory and the experimental data reproduce previous studies [1,2], showing an increasing loss with increasing sweep rate, followed by a universal plateau. This behavior corresponds to the regime of strong TLS relaxation between consecutive transitions, which can thus be considered independent [1,2]. In our study we explore a regime of high sweep rates, larger than the TLS relaxation rate, for which the coherent evolution during several Landau-Zener transitions has to be considered. We show that due to interference effects, the resonator loss decreases again in this regime. In contrast to the low dielectric loss at zero sweep rate, arising from saturation of photon absorption by TLSs, the low loss in the high sweep rate regime is a consequence of a reduced photon absorption probability due to the coherent evolution of the TLS state over many Landau-Zener transitions. The possibility to extend this physics in order to dynamically decouple TLSs from a qubit will be discussed. 1. M. S. Khalil, S. Gladchenko, M. J. A. Stoutimore, F. C. Wellstood, A. L. Burin, and K. D. Osborn, Phys. Rev. B 90, 100201(R) (2014). 2. A. L. Burin, M. S. Khalil, and K. D. Osborn, Phys. Rev. Lett. 110, 157002 (2013).

The investigation of non-equilibrium disordered quantum systems with novel experimental techniques have resulted in fundamentally new insights of their dynamics. In particular, the importance of nuclear spins as surprisingly active degrees of freedom at ultra-low temperatures has been revealed in recent measurements on amorphous solids. These new findings are of high relevance to many quantum devices, like quantum dots, qubits, SQUIDs, nano-mechanical systems and quantum limited amplifiers. We present the experimental evidence for a nuclear spin driven dynamics in non-equilibrium disordered quantum systems and discuss a possible theoretical framework for such a mechanism.

Amorphous solids manifest puzzling effects of mysterious degrees of freedom that give rise to a heat capacity and phonon scattering in great excess over what would be expected for a solid that has a unique vibrational ground state. Of particular conceptual importance is the apparent near universality of phonon scattering in amorphous solids made by quenching a liquid. To rationalise this universality, scale-free scenarios have been proposed that either hinge on there being long-range interactions between bare structural degrees of freedom or that invoke long-range criticality stemming from the emergence of marginally stable vibrational modes. In a contrasting, local scenario, the puzzling low-temperature degrees of freedom are, instead, weakly-interacting, strongly anharmonic degrees of freedom each of which involves the motion of a few hundred particles. In this scenario, the universality of phonon scattering comes about because the characteristic energy scale of the local anharmonic resonances and the strength of their interaction with phonons are both set by the glass transition temperature $T_g$, while their concentration is set by the cooperativity size $\xi$ for dynamics at $T_g$. The nanoscopic length $\xi$ is manifested in vibrational excitations of the spatial boundary of the resonances, which underlie the so-called Boson peak, and very deep, topological midgap electronic states in glassy semiconductors, which are implicated in a number of strange optoelectronic phenomena in amorphous chalcogenides. I discuss the merits of the above scenarios when confronted with experimental data.

Recently, there's renewed interest in the two TLS model as an explanation for the phenomena observed in experiments (M Schechter et al, 2018). We propose a new realization of the two-TLS model, based on ab-initio calculation and realistic interpretation of what is the mechanism generating the two TLSs (inversion symmetry breaking), as well as an actual simulation of the tunneling entity. We answer the question of "what tunnels" (Anderson et al, 1972), and suggest what (probable) tunneling systems may be found in real systems.

TBA

I will describe the experiments on transport of heat mediated by a superconducting qubit between two mesoscopic heat baths at different temperatures. I will also present theoretical models to account for the observations.

Low-frequency noise strongly impairs the performance of a variety of devices that are based on superconducting quantum circuits. This includes not only superconducting quantum interference devices (SQUIDs) but also superconducting qubits, low-temperature particle detectors and microwave resonators. However, in contrast to its frequent appearance, the origin and the general properties of low-frequency noise are often not well understood. This applies in particular to low-frequency excess flux noise which has been puzzling the community for more than 30 years. It limits, for example, the coherence times of flux or phase qubits and makes SQUID based high-precision measurements of low-frequency signals rather challenging. Recent experiments indicate that low-frequency excess flux noise in Josephson junction based superconducting quantum devices originates from the random reversal of interacting spins located at the device surface or in the interface between the substrate and the device. But even though this explanation presently proves more and more to be generally correct, the physical nature of these spins, i.e. their origin as well as their interaction mechanisms, has not yet been fully resolved and many open questions remain. In this talk, we will summarize the present knowledge of the community about the origin and the various properties of low-frequency excess flux noise. We will further discuss our most recent measurements which we performed to address several open question related to low-frequency excess flux noise. This includes, for example, the role of nuclear spins and surface adsorption layers as well as the dependencies of low-frequency excess flux noise on device geometry, device type and materials used for device fabrication. We will also show that these dependencies somehow facilitates to engineer the shape of magnetic flux noise spectra and thus to experimentally modify key properties of superconducting quantum devices such as coherence or measurement times.

Tunneling systems as well as qubits are typically suffering from various noise sources. A single noise source can induce relaxation and / or dephasing depending on how it interacts with the tunneling system or qubit. The dephasing influence of a pure dephasing noise and a independent noise source inducing relaxation is not simply additive. Surprisingly, a strong pure dephasing noise can also suppress partially relaxation from another noise source. These effects might severely complicate the identification of noise sources in qubits and tunneling systems.