pith. sign in

arxiv: 1907.00263 · v1 · pith:L6HLOQM5new · submitted 2019-06-29 · 🧬 q-bio.NC · cond-mat.supr-con· cs.NE

A Power Efficient Artificial Neuron Using Superconducting Nanowires

Emily Toomey , Ken Segall , Karl K. Berggren This is my paper

Pith reviewed 2026-05-25 12:32 UTC · model grok-4.3

classification 🧬 q-bio.NC cond-mat.supr-concs.NE
keywords superconducting nanowiresartificial neuronspiking neural networksneuromorphic hardwarelow-power electronicsvariable inductive synapse
0
0 comments X

The pith

Superconducting nanowires generate spiking behavior like biological neurons through their intrinsic nonlinearity.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper proposes using superconducting nanowires to build an artificial neuron. Two coupled nanowires produce spiking via their built-in nonlinearity, and circuit simulations show this matches several traits of biological neurons. A synapse design exploits the nanowire inductance nonlinearity to provide both excitatory and inhibitory signals while supporting direct fanout to multiple targets. The approach is presented as competitive in energy use with existing hardware for spiking networks.

Core claim

Building on an architecture first proposed for Josephson junctions, we rely on the intrinsic nonlinearity of two coupled nanowires to generate spiking behavior, and use electrothermal circuit simulations to demonstrate that the nanowire neuron reproduces multiple characteristics of biological neurons. Furthermore, by harnessing the nonlinearity of the superconducting nanowire's inductance, we develop a design for a variable inductive synapse capable of both excitatory and inhibitory control. We demonstrate that this synapse design supports direct fanout, a feature that has been difficult to achieve in other superconducting architectures, and that the nanowire neuron's nominal energy性能 is 竞争的

What carries the argument

Intrinsic nonlinearity of two coupled superconducting nanowires, which produces spiking through electrothermal circuit dynamics.

If this is right

  • The nanowire neuron reproduces multiple characteristics of biological neurons in the simulations.
  • The variable inductive synapse provides both excitatory and inhibitory control.
  • The synapse design supports direct fanout to other elements.
  • The nominal energy performance is competitive with that of current technologies.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Such a neuron could serve as a building block for hardware versions of spiking neural networks that operate at lower power than software implementations.
  • The fanout feature may allow larger network sizes than other superconducting neuron designs.
  • Integration with existing superconducting circuits could create hybrid systems for specialized neuromorphic tasks.

Load-bearing premise

The electrothermal circuit simulations used to demonstrate spiking behavior and synapse function accurately reflect the physical dynamics and fabrication realities of actual superconducting nanowire devices.

What would settle it

Fabricate the nanowire neuron circuit and measure its output signals under input conditions to determine whether spiking occurs as the simulations predict.

Figures

Figures reproduced from arXiv: 1907.00263 by Emily Toomey, Karl K. Berggren, Ken Segall.

Figure 1
Figure 1. Figure 1: Relaxation oscillations in superconducting nanowires, which serve as the foundation of the [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Circuit simulations of the two-nanowire soma, where the two oscillators act analogously to the two ion channels in the simplified neuron model. (a) Input pulse, Iin = 4 µA. (b) Current through the loop inductor. (c) Current through the control nanowire. The control nanowire reduces the amount of counter-clockwise current circulating in the loop, allowing the main nanowire to fire again. (d) Current through… view at source ↗
Figure 3
Figure 3. Figure 3: Firing threshold of the two-nanowire neuron. (a) Peak output voltage as a function of input current under a constant bias (Ibias = 58.6 µA). The plot illustrates that the neuron does not spike until the input current exceeds 4 µA, at which point it fires with output voltages of the same amplitude. Inset shows the time domain voltage output of the neuron for different input currents. (b) Input to the neuron… view at source ↗
Figure 4
Figure 4. Figure 4: Refractory period of the two-nanowire neuron. (a) Response when there is sufficient time between two inputs to each elicit a separate spike. Parameters: Ibias = 58 µA, Iin = 6 µA, Δt = 4 ns. The pink dashed lines indicate the beginning of the rising edge of each pulse. (b) Response when there is insufficient time between two input pulses, causing the neuron to fire only once. Parameters are the same as in … view at source ↗
Figure 5
Figure 5. Figure 5: Near-coincidence detection of pulses. Output voltage of the neuron when the time between successive input pulses Δt is a) 5 ns b) 3 ns. The pink dashed lines indicate the rising edge of each pulse. Two pulses must be in rapid succession in order to fire the under-biased neuron, demonstrating that it may be used for near-coincidence detection of input pulses. Parameters: Iin = 4.6 µA, Ibias = 57.62 µA [PIT… view at source ↗
Figure 6
Figure 6. Figure 6: Effect of bias current on spiking frequency. [PITH_FULL_IMAGE:figures/full_fig_p007_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Parabolic bursting in the two-nanowire neuron. (a) Output voltage of the two-nanowire neuron when the bias is coupled to a weak sinusoidal drive (f = 50 MHz, Iac = 4 µA). The red dashed curve indicates the sinusoidal drive, shifted in the y-axis for clarity. (b) The inverse of the time between adjacent peaks shows that the time difference follows a parabolic form. Parameters: Ibias = 59 µA, Iin = 6 µA. 3.6… view at source ↗
Figure 8
Figure 8. Figure 8: A superconducting transmission line as an axon. [PITH_FULL_IMAGE:figures/full_fig_p009_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: illustrates the circuit schematic of an inductive synapse that may be implemented in the nanowire neuron. Similar to the slow release of neurotransmitters in response to an action potential, the inductive synapse relies on the slow charging of a large inductor in response to the nanowire neuron’s more rapid voltage spikes. The energy stored in the large synapse inductor is then discharged as current into t… view at source ↗
Figure 10
Figure 10. Figure 10: Downstream control of the inductive synapse. (a) Excitatory control. Parameters: Ibias,main = 59 µA, Rseries = 14 Ω, Rsyn,1 = 40 Ω, Lsyn = 0.265 µH, Rsyn,2 = 40 Ω, Ibias,target = 57.17 µA, Iin = 4.6 µA. (b) Inhibitory control. Parameters Ibias,main = -58.6 µA, Rseries = 24 Ω, Rsyn,1 = 45 Ω, Lsyn = 0.23 µH, Rsyn,2 = 40 Ω, Ibias,target = 57.68 µA, Iin = 4.6 µA. For both cases: Panel (i) displays the output … view at source ↗
Figure 11
Figure 11. Figure 11: Modulating the inductive synapse. (a) Circuit schematic of the inductive synapse with a high￾inductance nanowire and ideal current source placed in parallel. L1, L2 << Lnanowire, Lsyn. Rseries, out has been added to further prevent backaction from the target neuron. (b) Simulation of the current through Rseries,out in an inhibitory synapse as a function of different modulation currents. Spikes represent b… view at source ↗
Figure 12
Figure 12. Figure 12: Fanout of nanowire neuron with a tunable inductive synapse. [PITH_FULL_IMAGE:figures/full_fig_p012_12.png] view at source ↗
read the original abstract

With the rising societal demand for more information-processing capacity with lower power consumption, alternative architectures inspired by the parallelism and robustness of the human brain have recently emerged as possible solutions. In particular, spiking neural networks (SNNs) offer a bio-realistic approach, relying on pulses analogous to action potentials as units of information. While software encoded networks provide flexibility and precision, they are often computationally expensive. As a result, hardware SNNs based on the spiking dynamics of a device or circuit represent an increasingly appealing direction. Here, we propose to use superconducting nanowires as a platform for the development of an artificial neuron. Building on an architecture first proposed for Josephson junctions, we rely on the intrinsic nonlinearity of two coupled nanowires to generate spiking behavior, and use electrothermal circuit simulations to demonstrate that the nanowire neuron reproduces multiple characteristics of biological neurons. Furthermore, by harnessing the nonlinearity of the superconducting nanowire's inductance, we develop a design for a variable inductive synapse capable of both excitatory and inhibitory control. We demonstrate that this synapse design supports direct fanout, a feature that has been difficult to achieve in other superconducting architectures, and that the nanowire neuron's nominal energy performance is competitive with that of current technologies.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper proposes an artificial neuron based on two coupled superconducting nanowires that exploits their intrinsic nonlinearity to generate spiking behavior, as shown through electrothermal circuit simulations that reproduce several biological neuron characteristics. It further introduces a variable inductive synapse design leveraging nanowire inductance nonlinearity for excitatory and inhibitory control, demonstrating direct fanout capability and claiming competitive energy performance relative to existing technologies.

Significance. If the simulated behaviors accurately map to fabricated devices, this could provide a low-power, scalable hardware platform for spiking neural networks in neuromorphic computing, with the fanout-supporting synapse addressing a noted limitation in other superconducting approaches. The work credits the use of electrothermal simulations to explore multiple neuron-like traits and synapse functionality without introducing free parameters in the core claims.

major comments (2)
  1. [Abstract] Abstract (and simulation results description): The central claim that the nanowire neuron reproduces multiple biological characteristics and that the synapse supports fanout with competitive energy rests entirely on electrothermal circuit simulations; however, no experimental validation against measured nanowire devices, no parameter sensitivity analysis, and no Monte-Carlo studies of fabrication variations (e.g., critical-current inhomogeneity or thermal diffusion effects) are reported, making the mapping from numerics to physical behavior untested and load-bearing for the claims.
  2. [Abstract] Abstract (simulation paragraph): The weakest assumption—that the electrothermal model captures all dominant physical effects under realistic tolerances—is not addressed with any robustness checks, which directly undermines the translation of the reported spiking behavior and energy performance to hardware.
minor comments (2)
  1. Notation for the inductive synapse circuit elements could be clarified with an explicit diagram or equation set to distinguish excitatory vs. inhibitory modes.
  2. The manuscript would benefit from a brief comparison table of simulated energy per spike against the referenced current technologies.

Simulated Author's Rebuttal

2 responses · 2 unresolved

We thank the referee for the constructive comments on our manuscript. The work is a simulation-based proposal using established electrothermal models, and we address the concerns regarding validation and robustness below.

read point-by-point responses

  1. Referee: [Abstract] Abstract (and simulation results description): The central claim that the nanowire neuron reproduces multiple biological characteristics and that the synapse supports fanout with competitive energy rests entirely on electrothermal circuit simulations; however, no experimental validation against measured nanowire devices, no parameter sensitivity analysis, and no Monte-Carlo studies of fabrication variations (e.g., critical-current inhomogeneity or thermal diffusion effects) are reported, making the mapping from numerics to physical behavior untested and load-bearing for the claims.

    Authors: The manuscript presents a simulation study to explore the feasibility of the proposed nanowire neuron and synapse using a standard electrothermal circuit model with parameters drawn from published measurements on superconducting nanowire devices. We agree that experimental validation, parameter sensitivity analysis, and Monte-Carlo studies of fabrication variations would strengthen the mapping to hardware. These elements were not included because the scope was limited to demonstrating core spiking behaviors and fanout capability within the model. We will revise the manuscript to add an explicit discussion of model assumptions, parameter ranges explored, and the need for future experimental confirmation. revision: partial

  2. Referee: [Abstract] Abstract (simulation paragraph): The weakest assumption—that the electrothermal model captures all dominant physical effects under realistic tolerances—is not addressed with any robustness checks, which directly undermines the translation of the reported spiking behavior and energy performance to hardware.

    Authors: The electrothermal model incorporates the dominant effects of Joule heating, thermal diffusion, and superconducting transition as described in prior literature on nanowire devices. While we did not report dedicated robustness checks such as systematic sensitivity sweeps or Monte-Carlo runs in the original submission, the spiking and synaptic behaviors were observed consistently across the parameter sets used. We will add a dedicated subsection in the revised manuscript discussing model limitations, the range of parameters tested, and the implications for hardware translation. revision: yes

standing simulated objections not resolved
  • Experimental validation against fabricated nanowire devices
  • Monte-Carlo studies of fabrication variations (critical-current inhomogeneity, thermal effects)

Circularity Check

0 steps flagged

No circularity: simulation outputs of proposed circuit design

full rationale

The paper proposes a nanowire neuron architecture and reports outputs from electrothermal circuit simulations that reproduce biological neuron traits and synapse behavior. No step reduces a claimed prediction or first-principles result to its own inputs by definition, fitted-parameter renaming, or self-citation chains. The architecture is presented as building on prior Josephson-junction work without invoking load-bearing uniqueness theorems from the same authors. Claims remain independent numerical demonstrations rather than tautological redefinitions.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Abstract-only review provides no explicit free parameters, axioms, or invented entities; the demonstration depends on the unstated assumption that standard superconducting nanowire models suffice for the proposed circuits.

axioms (1)
  • domain assumption Electrothermal models of superconducting nanowires are sufficiently accurate to predict spiking and synaptic behavior in the proposed circuit.
    All performance claims rest on these simulations matching physical reality.

pith-pipeline@v0.9.0 · 5752 in / 1299 out tokens · 40876 ms · 2026-05-25T12:32:18.703298+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

37 extracted references · 37 canonical work pages

  1. [1]

    Neural systems engineering,

    S. Furber and S. Temple, “Neural systems engineering,” J. R. Soc. Interface, vol. 4, no. 13, pp. 193– 206, Apr. 2007

    work page 2007

  2. [2]

    A 4-fJ/Spike Artificial Neuron in 65 nm CMOS Technology,

    I. Sourikopoulos et al., “A 4-fJ/Spike Artificial Neuron in 65 nm CMOS Technology,” Front. Neurosci., vol. 11, 2017

    work page 2017

  3. [3]

    Magnetic skyrmion-based artificial neuron device,

    S. Li, W. Kang, Y. Huang, X. Zhang, Y. Zhou, and W. Zhao, “Magnetic skyrmion-based artificial neuron device,” Nanotechnology, vol. 28, no. 31, p. 31LT01, Jul. 2017

    work page 2017

  4. [4]

    Simulation of a memristor-based spiking neural network immune to device variations,

    D. Querlioz, O. Bichler, and C. Gamrat, “Simulation of a memristor-based spiking neural network immune to device variations,” in The 2011 International Joint Conference on Neural Networks, 2011, pp. 1775–1781

    work page 2011

  5. [5]

    Memristor-based neural networks,

    A. Thomas, “Memristor-based neural networks,” J. Phys. Appl. Phys., vol. 46, no. 9, p. 093001, Feb. 2013

    work page 2013

  6. [6]

    Josephson junction simulation of neurons,

    P. Crotty, D. Schult, and K. Segall, “Josephson junction simulation of neurons,” Phys. Rev. E, vol. 82, no. 1, p. 011914, Jul. 2010

    work page 2010

  7. [7]

    Ultralow power artificial synapses using nanotextured magnetic Josephson junctions,

    M. L. Schneider et al., “Ultralow power artificial synapses using nanotextured magnetic Josephson junctions,” Sci. Adv., vol. 4, no. 1, p. e1701329, Jan. 2018

    work page 2018

  8. [8]

    Frequency Pulling and Mixing of Relaxation Oscillations in Superconducting Nanowires,

    E. Toomey, Q.-Y. Zhao, A. N. McCaughan, and K. K. Berggren, “Frequency Pulling and Mixing of Relaxation Oscillations in Superconducting Nanowires,” Phys. Rev. Appl., vol. 9, no. 6, p. 064021, Jun. 2018

    work page 2018

  9. [9]

    Kinetic-inductance-limited reset time of superconducting nanowire photon counters,

    A. J. Kerman et al., “Kinetic-inductance-limited reset time of superconducting nanowire photon counters,” Appl. Phys. Lett., vol. 88, no. 11, p. 111116, Mar. 2006

    work page 2006

  10. [10]

    The Hodgkin–Huxley Equations,

    G. B. Ermentrout and D. H. Terman, “The Hodgkin–Huxley Equations,” in Mathematical Foundations of Neuroscience, vol. 35, New York, NY: Springer New York, 2010, pp. 1–28

    work page 2010

  11. [11]

    Electrothermal feedback in superconducting nanowire single-photon detectors,

    A. J. Kerman, J. K. W. Yang, R. J. Molnar, E. A. Dauler, and K. K. Berggren, “Electrothermal feedback in superconducting nanowire single-photon detectors,” Phys. Rev. B, vol. 79, no. 10, p. 100509, Mar. 2009

    work page 2009

  12. [12]

    A Superconducting Nanowire can be Modeled by Using SPICE,

    K. K. Berggren et al., “A Superconducting Nanowire can be Modeled by Using SPICE,” Supercond. Sci. Technol., 2018

    work page 2018

  13. [13]

    Which model to use for cortical spiking neurons?,

    E. M. Izhikevich, “Which model to use for cortical spiking neurons?,” IEEE Trans. Neural Netw., vol. 15, no. 5, pp. 1063–1070, Sep. 2004

    work page 2004

  14. [14]

    Topological and phenomenological classification of bursting oscillations,

    R. Bertram, M. J. Butte, T. Kiemel, and A. Sherman, “Topological and phenomenological classification of bursting oscillations,” Bull. Math. Biol., vol. 57, no. 3, pp. 413–439, Jan. 1995

    work page 1995

  15. [15]

    Bursting oscillations in an excitable membrane model,

    J. Rinzel, “Bursting oscillations in an excitable membrane model,” in Ordinary and Partial Differential Equations, 1985, pp. 304–316

    work page 1985

  16. [16]

    Elliptic and parabolic bursting in a digital silicon neuron model,

    T. Nanami, K. Aihara, and T. Kohno, “Elliptic and parabolic bursting in a digital silicon neuron model,” presented at the 2016 International Symposium on Nonlinear Theory and Its Applications, Yugawara, Japan, 2016, p. 4

    work page 2016

  17. [17]

    Spiking and Bursting in Josephson Junction,

    S. K. Dana, D. C. Sengupta, and C.- Hu, “Spiking and Bursting in Josephson Junction,” IEEE Trans. Circuits Syst. II Express Briefs, vol. 53, no. 10, pp. 1031–1034, Oct. 2006

    work page 2006

  18. [18]

    Parabolic bursting revisited,

    C. Soto-Treviño, N. Kopell, and D. Watson, “Parabolic bursting revisited,” J. Math. Biol., vol. 35, no. 1, pp. 114–128, Nov. 1996

    work page 1996

  19. [19]

    The demonstration and manipulation of a circadian rhythm in a single neuron,

    F. Strumwasser, “The demonstration and manipulation of a circadian rhythm in a single neuron,” in Circadian clocks, Amsterdam: North-Holland Publishing Co., 1965, pp. 442–462

    work page 1965

  20. [20]

    Spontaneous Activity in Isolated Somata of Aplysia Pacemaker Neurons,

    B. O. Alving, “Spontaneous Activity in Isolated Somata of Aplysia Pacemaker Neurons,” J. Gen. Physiol., vol. 51, no. 1, p. 29, Jan. 1968

    work page 1968

  21. [21]

    Single-photon imager based on a superconducting nanowire delay line,

    Q.-Y. Zhao et al., “Single-photon imager based on a superconducting nanowire delay line,” Nat. Photonics, vol. 11, no. 4, pp. 247–251, Apr. 2017

    work page 2017

  22. [22]

    A distributed electrical model for superconducting nanowire single photon detectors,

    Q.-Y. Zhao, D. F. Santavicca, D. Zhu, B. Noble, and K. K. Berggren, “A distributed electrical model for superconducting nanowire single photon detectors,” Appl. Phys. Lett., vol. 113, no. 8, p. 082601, Aug. 2018

    work page 2018

  23. [23]

    The action potential in mammalian central neurons,

    B. P. Bean, “The action potential in mammalian central neurons,” Nat. Rev. Neurosci., vol. 8, p. 451, Jun. 2007

    work page 2007

  24. [24]

    Comparison of Cortico-Cortical and Cortico-Collicular Signals for the Generation of Saccadic Eye Movements,

    S. Ferraina, M. Paré, and R. H. Wurtz, “Comparison of Cortico-Cortical and Cortico-Collicular Signals for the Generation of Saccadic Eye Movements,” J. Neurophysiol., vol. 87, no. 2, pp. 845– 858, Feb. 2002

    work page 2002

  25. [25]

    Tunable superconducting nanoinductors,

    A. J. Annunziata et al., “Tunable superconducting nanoinductors,” Nanotechnology, vol. 21, no. 44, p. 445202, Oct. 2010

    work page 2010

  26. [26]

    A Superconducting-Nanowire Three-Terminal Electrothermal Device,

    A. N. McCaughan and K. K. Berggren, “A Superconducting-Nanowire Three-Terminal Electrothermal Device,” Nano Lett., vol. 14, no. 10, pp. 5748–5753, Oct. 2014

    work page 2014

  27. [27]

    A million spiking-neuron integrated circuit with a scalable communication network and interface,

    P. A. Merolla et al., “A million spiking-neuron integrated circuit with a scalable communication network and interface,” Science, vol. 345, no. 6197, pp. 668–673, Aug. 2014

    work page 2014

  28. [28]

    Large-scale neuromorphic computing systems,

    S. Furber, “Large-scale neuromorphic computing systems,” J. Neural Eng., vol. 13, no. 5, p. 051001, Aug. 2016

    work page 2016

  29. [29]

    A nanocryotron comparator can connect single-flux-quantum circuits to conventional electronics,

    Q.-Y. Zhao, A. N. McCaughan, A. E. Dane, K. K. Berggren, and T. Ortlepp, “A nanocryotron comparator can connect single-flux-quantum circuits to conventional electronics,” Supercond. Sci. Technol., vol. 30, no. 4, p. 044002, 2017

    work page 2017

  30. [30]

    Circuit designs for superconducting optoelectronic loop neurons,

    J. M. Shainline et al., “Circuit designs for superconducting optoelectronic loop neurons,” J. Appl. Phys., vol. 124, no. 15, p. 152130, Oct. 2018

    work page 2018

  31. [31]

    Fluxonic Processing of Photonic Synapse Events,

    J. M. Shainline, “Fluxonic Processing of Photonic Synapse Events,” ArXiv190402807 Cs, Apr. 2019

    work page 2019

  32. [32]

    Picosecond superconducting single-photon optical detector,

    G. N. Gol’tsman et al., “Picosecond superconducting single-photon optical detector,” Appl. Phys. Lett., vol. 79, no. 6, pp. 705–707, Aug. 2001

    work page 2001

  33. [33]

    Single-Photon Detectors Based on Ultranarrow Superconducting Nanowires,

    F. Marsili et al., “Single-Photon Detectors Based on Ultranarrow Superconducting Nanowires,” Nano Lett., vol. 11, no. 5, pp. 2048–2053, May 2011

    work page 2048

  34. [34]

    Nanoscale superconducting memory based on the kinetic inductance of asymmetric nanowire loops,

    A. Murphy, D. V. Averin, and A. Bezryadin, “Nanoscale superconducting memory based on the kinetic inductance of asymmetric nanowire loops,” New J. Phys., vol. 19, no. 6, p. 063015, 2017

    work page 2017

  35. [35]

    A compact superconducting nanowire memory element operated by nanowire cryotrons,

    Q.-Y. Zhao et al., “A compact superconducting nanowire memory element operated by nanowire cryotrons,” Supercond. Sci. Technol., vol. 31, no. 3, p. 035009, 2018

    work page 2018

  36. [36]

    A kinetic- inductance-based superconducting memory element with shunting and sub-nanosecond write times,

    A. N. McCaughan, E. A. Toomey, M. Schneider, K. K. Berggren, and S. W. Nam, “A kinetic- inductance-based superconducting memory element with shunting and sub-nanosecond write times,” Supercond. Sci. Technol., 2018

    work page 2018

  37. [37]

    Bridging the Gap Between Nanowires and Josephson Junctions: A Superconducting Device Based on Controlled Fluxon Transfer,

    E. Toomey, M. Onen, M. Colangelo, B. A. Butters, A. N. McCaughan, and K. K. Berggren, “Bridging the Gap Between Nanowires and Josephson Junctions: A Superconducting Device Based on Controlled Fluxon Transfer,” Phys. Rev. Appl., vol. 11, no. 3, p. 034006, Mar. 2019

    work page 2019