AVS2018 Session MP+EM+NS-TuM: High Coherence Qubits for Quantum Computing

Tuesday, October 23, 2018 8:00 AM in Room 203A

Tuesday Morning

Start	Invited?	Item
8:00 AM		MP+EM+NS-TuM-1 MBE Grown Nitride Superconductors for Quantum Circuits Christopher Richardson, Ashish Alexander, Christopher Weddle (Laboratory for Physical Sciences) Low microwave loss superconducting capacitors and inductors are critical circuit components of superconducting qubits. For transmon qubits, the ability to make high-quality planar resonators is an essential part of fabricating highly coherent qubits. Plasma assisted Molecular beam epitaxy (PAMBE) is used to grow niobium titanium nitride alloys (Nb_xTi_1-xN) directly on silicon (111) wafers. Using a structure first approach to design optimization, the structural, surface topology, chemical characteristics, and superconducting critical temperature of these films are used for optimization of the growth conditions before resonators are fabricated and tested. Here focus will be on the optimization of PAMBE-TiN films grown under slightly nitrogen rich conditions and high growth temperatures. Using films grown on high resistivity wafers, resonators are fabricated from coplanar waveguides with a narrow 6-µm wide center conductor and nominal 500-nm deep trench etch. Cryogenic testing at 100mK demonstrate low microwave loss that is evident from measured internal quality factors that are over 1M in the single photon regime and approach 10M at high powers. The motivation of using PAMBE to grow superconductors and the favorable comparison with resonators made from leading films synthesized with sputter deposition will also be discussed.
8:20 AM		MP+EM+NS-TuM-2 Towards Improved Coherence Times in Transmon Qubits Sam Stanwyck (Rigetti Computing) The depth of the circuit a quantum computer can perform depends directly on the coherence times of its qubits. There are many sources of decoherence in superconducting qubits, and identifying and minimizing dominant sources is a critical step in improving the performance of quantum computers. By measuring the internal quality factor of resonators and deliberately coupling to different loss mechanisms, we identify dominant sources of resonator loss in our systems, as well fabrication process changes to ameliorate these losses. Additionally, defects and materials present on the chip surface are correlated with process changes and coherence metrics.
8:40 AM	Invited	MP+EM+NS-TuM-3 Design and Fabrication for High Coherence Quantum Circuits David Pappas, Xian Wu, Russell Lake, Mustafa Bal, Junling Long, Corey Rae McRae, Hsiang-Sheng Ku (National Institute of Standards and Technology (NIST)) In this talk we focus on achieving high coherence in multi-component quantum circuits [1,13]. We will discuss geometric and electrical design strategies that mitigate energy loss while maintaining sufficient coupling to the qubit. Materials considerations -including dielectric losses in the substrate and various interfaces -play a central role in the implementation of these circuits. We will present a summary of our studies of the various participation factors and processing techniques to reduce dielectric loss in the capacitance of the qubits and resonators for readout and coupling. We also review our methods of integration for the key nonlinear component, the overlap tunnel junctions. In particular, techniques for achieving smooth surfaces for the junctions in a back-end process will be shown. [1] X. Wu, et al., Appl. Phys. Lett. 111, 032602 (2017); https://doi.org/10.1063/1.4993937 [2] D.P. Pappas, Appl. Phys. Lett. 112, 182601 (2018); doi: 10.1063/1.5027104 [3] N.T. Bronn, et al., Quantum Sci. Technol. 3 (2018) 024007. [4] P. Kumar, et al., Phys. Rev. Appl. 6, 041001 (2016). [5] J. Braumuller, et al., Appl. Phys. Lett. 108, 032601 (2016). [6] J.B. Chang, et al.,Appl. Phys. Lett. 103, 012602 (2013). [7] M. Sandberg, et al., Appl. Phys. Lett. 102, 072601 (2013). [8] M. R. Sandberg, et al., Appl. Phys. Lett. 100, 082602 (2012). [9] M. R. Vissers, et al., Appl. Phys. Lett. 101, 022601 (2012). [10] M. Sandberg, et al.,Appl. Phys. Lett. 100, 262605 (2012). [11] D. P. Pappas, et al., IEEE Trans. Appl. Supercon., Vol. 21, No. 3, June 2011. [12] D. S. Wisbey, et al., J. Appl. Phys. 108, 093918 (2010). [13] J. M. Martinis, PRL 95, 210503 (2005). View Supplemental Document (pdf)
9:20 AM		MP+EM+NS-TuM-5 Effect of Surface Treatment on Superconducting Qubit Coherence Bradley Christensen (University of Wisconsin-Madison); Pradeep Kumar (University of Wisconsin - Madison); J.J. Nelson, Yebin Liu, Andrew Ballard, Britton Plourde (Syracuse University); Robert McDermott (University of Wisconsin - Madison) Superconducting qubits are an attractive candidate for quantum information processing in the solid state . The fidelity of two-qubit gates for superconducting qubits is one of the more challenging limitations toward scalable quantum computing. A promising approach to perform these gates uses flux-tunable qubits to bias the qubit pairs into resonance to perform the necessary entangling operations. While this approach has many advantages over competing techniques, there are still significant issues that limit the fidelity of the gates. For example, since the two-qubit gate requires flux-biasing of a qubit, this also necessarily requires one of the qubits to operate at a flux-sensitive point, and as such, 1/f flux noise will restrict the possible gate fidelity. In addition to flux noise, flux-tunable qubits also suffer from microscopic two-level system (TLS) defects that reside in the high field areas on the qubit capacitor pad. These TLS cause an enhanced decay through both resonant interactions with the qubit and Landau-Zener transitions as the qubit frequency is biased through a TLS. While one could perform spectroscopy of the TLS to map out the inoperable space, this becomes an inefficient solution for large scale systems as the TLS resonance frequencies are not stable, and will significantly drift over time. Recent experiments on SuperconductingQUantum Interference Devices (SQUID) point to adsorbed molecular O₂ as the dominant contributor to magnetism in superconducting thin ﬁlms, and demonstrate that improvements in the sample vacuum environment lead to signiﬁcant reductions in surface spin susceptibility and magnetic ﬂux noise power. Furthermore, TLS defects have been shown to reside in surface oxides and interfaces, where the TLS dipole couples to the qubit electric field, and experiments on microwave resonators have shown that high-temperature annealing can yield a reduction in surface TLS defects. Here, we present our results on improving the vacuum environment of superconducting qubits with an ultra-high vacuum (UHV) bake to remove the adsorbates. We measure flux noise power spectral densities (PSD) using Ramsey-based, CPMG filtering, and dressed-dephasing techniques, allowing the flux PSD measurements to span 10 decades. Furthermore, by measuring qubit lifetime as a function of frequency (swap spectroscopy), we can map out the coupling strength, lifetime, and density of the TLS defects. We present a comparison of treated and untreated devices to demonstrate the improvement to qubit coherence through a UHV bake.
9:40 AM		MP+EM+NS-TuM-6 Metrology of Dielectric Loss using Lumped-Element Microwave Resonators Corey Rae McRae, Xian Wu, Mustafa Bal, Junling Long, Hsiang-Sheng Ku, David Pappas, Russell Lake (National Institute of Standards and Technology) Reducing the overall concentration of TLSs in dielectric materials remains at the forefront of materials research in quantum information science. In this work, we measure a lumped element resonator fabricated from a superconductor-dielectric-superconductor trilayer to determine the TLS loss of various dielectrics of interest in superconducting quantum computing. The deposition of the trilayer prior to fabrication allows control of the metal-dielectric interfaces, and the fabrication process is generalized so that resonators containing different dielectrics can be compared easily. This lithography method enables the measurement of trilayer capacitors and junctions that have been prepared entirely in situ in an ultrahigh vacuum environment. In future work, we will interrogate a new class of low-loss dielectrics grown with epitaxial methods using the measurement capabilities developed here.
10:00 AM		BREAK - Complimentary Coffee in Exhibit Hall
11:00 AM	Invited	MP+EM+NS-TuM-10 Direct Observation of Atomic Structure of Ultra Thin AlO_x Barriers in Al/AlO_x/Al Josephson Junctions for Quantum Devices Eva Olsson (Chalmers University of Technology, Gothenburg, Sweden) The atomic structure of tunnel barriers in Josephson junctions for quantum devices and the corresponding interfaces determine the properties of the junction. The thinnest region in the barrier of a junction will be the preferential tunneling channel for charge carriers and the highest current. The current increases exponentially with decreasing barrier thickness. As a consequence, a variation on the individual atom plane length scale results in inhomogeneity of the tunnel current across the barrier. There are several earlier experimental indirect indications that only a small fraction of the junction area is active. We are using high resolution annular dark field (ADF) scanning transmission electron microscopy (STEM) imaging to obtain high resolution (better than 1 Å) and high precision (better than 1 pm) information about the local atomic structure [1]. We use ADF STEM imaging to directly determine the thickness distribution along the oxide barrier in Al/AlO_x/Al Josephson junctions [2]. The barrier thickness is about 1-2 nm. The thickness distribution shows that less than 10% of the junction area dominates the electron tunneling. We also study the influence of oxygen pressure and oxidation time on the thickness distribution. In addition, we determine the atomic structure and coordination of Al atoms within the oxide barrier layer using electron energy loss spectroscopy and nanobeam electron diffraction [3]. A lower Al coordination is observed at the metal/oxide interface compared to the interior of the oxide barrier. We also study the structure of the interfaces between the Al contact and the substrate [4,5]. [1] A.B. Yankovich, R. Verre, E. Olsen, A.E.O. Persson, V. Trinh, G. Dovner, M. Käll and E. Olsson, ACS Nano 11 (2017) 4265. [2] L.J. Zeng, S. Nik, T. Greibe, P. Krantz, P. Delsing and E. Olsson, J. Phys. D: Appl. Phys.48 (2015) 395308 [3] L.J. Zeng, D.T. Tran, C.-W- Tai, G. Svensson and E. Olsson, Sci. Rep. 6 (2016) 29679. [4] L.J. Zeng, T. Greibe, S. Nik, C.M. Wilson, P. Delsing and E. Olsson, J. Appl. Phys. 113 (2013) 143905 [5] L.J. Zeng, P. Krantz, S. Nik, P. Delsing and E. Olsson, J. Appl. Phys. 117 (2015) 163915.
11:40 AM		MP+EM+NS-TuM-12 Metrology of Tunnel Junctions for Superconducting Qubits Russell Lake (National Institute of Standards and Technology (NIST)); Xian Wu, Hsiang-Sheng Ku, Junling Long, Mustafa Bal, Corey Rae McRae (National Institute of Standards and Technology (NIST) and University of Colorado Boulder); David Pappas (National Institute of Standards and Technology (NIST)) Superconducting tunnel junctions make up the key non-linear circuit component in many implementations of quantum electrical circuits, including superconducting qubits. Therefore, controllable fabrication of superconducting junctions has taken a central role in the realization of quantum computers. In this talk we discuss fabrication and characterization of a wafer-scale process for nanoscale superconducting tunnel junctions (Al-AlOx-Al) [1]. We present the distribution of normal-state resistances across a wafer for different junction sizes. We have applied an analytical method of accounting for the current crowding in the junction leads [2] in order to give accurate predictions of the supercurrent from the room-temperature raw data. These corrected resistances can be input into the Ambegaokar-Baratoff formula to predict the critical current of the tunnel junctions in the superconducting state [3], and the corresponding non-linear effective inductance. These results are immediately relevant to the task of qubit frequency allocation in multi-qubit systems. [1] Appl. Phys. Lett. 111, 032602 (2017); https://doi.org/10.1063/1.4993937 [2] J. Appl. Phys. 105, 094503 (2009); https://doi.org/10.1063/1.3122503 [3] Phys. Rev. Lett. 10, 486 (1963) and 11, 104 (1963); https://doi.org/10.1103/PhysRevLett.10.486