E ff ects of Radioactivity on Superconducting Quantum Bits Laura - - PowerPoint PPT Presentation

Effects of Radioactivity on Superconducting Quantum Bits

Laura Cardani for the DEMETRA Collaboration Istituto Nazionale di Fisica Nucleare - Roma 09/09/2019 TAUP , Toyama, Japan

Superconducting QUBITS

Laura Cardani, INFN - Roma TAUP (Toyama, Japan)

• Based on simple circuital elements (capacitor,

inductor…)

• Simple to fabricate and operate
• Fast gate time, high fidelity
• But poor coherence time

J.Gambetta, https://www.nature.com/articles/s41534-016-0004-0

Qubit: any two level system (many technologies proposed ) One of the most promising implementation: superconducting circuits

• 2
• V (t)

= V0 cos ω01t

Very short π-pulse time

Coherence Time

Laura Cardani, INFN - Roma TAUP (Toyama, Japan) Time in which a qubit retains its quantum behaviour

• 3
• Dielectric two levels system
• Paramagnetic molecules at the interface
• Vortices trapped in the superconductor
• ….
• Quasiparticles

Coherence Time

Laura Cardani, INFN - Roma TAUP (Toyama, Japan)

• Dielectric two levels system
• Paramagnetic molecules at the interface
• Vortices trapped in the superconductor
• ….
• Quasiparticles

Time in which a qubit retains its quantum behaviour

• 4
• Quasiparticles (“broken Cooper pairs”) can be viewed as free electrons
• They are dissipative
• They lead to decoherence
• Coherence time of 1ms —> 1QP for 109 Cooper pairs

Signatures of Quasiparticles

Laura Cardani, INFN - Roma TAUP (Toyama, Japan) A “steady population” worsen the performance (short coherence, low Q) Non equilibrium QP results in “bursts”

• 5

The higher the number of bursts, the shorter the coherence time

Pop 2018 doi:10.1103/PhysRevLett.121.117001

Radioactivity

Laura Cardani, INFN - Roma TAUP (Toyama, Japan)

• 6

Why do we expect radioactivity to produce quasiparticles?

• Qubits are very tiny —> almost

radioactive free

• But they are deposited on a much

wider substrate (Si or sapphire)

• Radioactivity hits the substrate producing

phonons

• Phonons travel in the substrate and hit the

qubit

• Whey they hit the qubit they produce QP bursts

For us rather obvious (N.Casali’s talk on New Techn. 4, Wed), not for qubit

DEMETRA experiment

Laura Cardani, INFN - Roma TAUP (Toyama, Japan)

• 7

Funded by INFN (Grant73/2018) Our goals:

• 1. Prove that qubits suffer from radioactivity
• 2. Demonstrate the key role of the substrate
• 3. Demonstrate that radioactivity suppression is achievable
• 4. Derive the impact qubit performance

DEMETRA experiment

Laura Cardani, INFN - Roma TAUP (Toyama, Japan)

• 8

Funded by INFN (Grant73/2018) Our goals:

• 1. Prove that qubits suffer from radioactivity
• 2. Demonstrate the key role of the substrate
• 3. Demonstrate that radioactivity suppression is achievable
• 4. Derive the impact qubit performance

Do QUBIT see radioactivity?

Laura Cardani, INFN - Roma TAUP (Toyama, Japan)

• 9

We measured the QP bursts in three superconducting circuits operated as KIDs. We faced a ThO source to the device.

• The closer the source, the higher the rate (up to x100)

Qubits will see environmental radioactivity!

DEMETRA experiment

Laura Cardani, INFN - Roma TAUP (Toyama, Japan)

• 10

Funded by INFN (Grant73/2018) Our goals:

• 1. Prove that qubits suffer from radioactivity
• 2. Demonstrate the key role of the substrate
• 3. Demonstrate that radioactivity suppression is achievable
• 4. Derive the impact qubit performance

Substrate or qubit itself?

Laura Cardani, INFN - Roma TAUP (Toyama, Japan)

• 11

Two hypotheses:

• If the radioactive interaction is in the qubit —> single burst
• If it is in the substrate —> bursts in all the 3 devices
• full response

1 2 3 4

DEMETRA experiment

Laura Cardani, INFN - Roma TAUP (Toyama, Japan)

• 12

Funded by INFN (Grant73/2018) Our goals:

• 1. Prove that qubits suffer from radioactivity
• 2. Demonstrate the key role of the substrate
• 3. Demonstrate that radioactivity suppression is achievable
• 4. Derive the impact qubit performance

Radioactivity Suppression

Laura Cardani, INFN - Roma TAUP (Toyama, Japan)

• 13
• Measurement in the above-ground laboratory of KIT

(Germany)

• Basic cleaning and repeat measurement in Roma

(Italy)

• Move Roma sample (and readout) in the Underground

LNGS (Italy)

• Bursts rate in Roma smaller than KIT (cleaning the sample made something)
• Bursts rate in LNGS much smaller than Roma/KIT
• Proved a suppression by more than x10

ΓB (mHz) 10 102

K R G G, no lead G + Th02

A B C

Impact on Intrinsic Performance

Laura Cardani, INFN - Roma TAUP (Toyama, Japan)

• 14
• How will the rate correlate with the qubit performance?
• KIT/Roma changed because of the set-up
• At LNGS, results better than in any other above-ground attempt!
• Effect of the source still under investigation

105 Qi @ ¯ n ⇡ 1 K R G G, no lead G + Th02

[kHz] f Frequency -

40 − 20 − 20 40

Amplitude [dB]

9 10 11 12 13 14 15 16

I [A.U.]

6.5 − 6 − 5.5 − 5 − 4.5 − 4 − 3.5 − 3 − 2.5 −

Q [A.U.]

1.5 − 1 − 0.5 − 0.5 1 1.5 2 2.5

• Measurements of quality factor
• (Proportional to coherence)

A B C

Conclusions and Perspectives

Laura Cardani, INFN - Roma TAUP (Toyama, Japan)

• 15

What next: 1) Understand the mechanisms relating radioactivity to coherence: 1) The energy distribution matters? 2) The time response matters? 2) Quantify: “how much” radioactivity suppression do we actually need for quantum processors? Our goals:

• 1. Prove that qubits suffer from radioactivity
• 2. Demonstrate the key role of the substrate
• 3. Demonstrate that radioactivity suppression is achievable
• 4. Derive the impact qubit performance
• M. Martinez et al.

Phys Rev Apply 2019

Thanks for the Attention!

Laura Cardani, INFN - Roma TAUP (Toyama, Japan)

• 16
• I. Colantoni

–

• L. Cardani, N. Casali, M. Clemenza, A. Cruciani, L. Gironi, S.

Pirro, C. Rusconi, M. Vignati

• M. Martinez
• T. Charpentier, L. Gruenhaupt, D. Gusenkova, F

. Henriques,

• M. Lagoin, I. Pop, F

. Valenti, W. Wernsdorfer, A. Ustinov

• G. Catelani

We welcome new collaborators!

QUANTUM BITS

Laura Cardani, INFN - Roma Low Radioactivity Techniques, Jaca (Spain)

α |0i + β |1i

Key feature: entanglement

Ψ = X

s1,...,sN

αs1,..,sN ψs1,..,sN

Quantum computer with n qubits: 2n-1 complex numbers Classical memory with n bits: string of n zeroes and ones Fundamental unit of information in a quantum computer Any two level quantum system

• 18

BUILDING SUPERCONDUCTING QUBITS

Laura Cardani, INFN - Roma Low Radioactivity Techniques, Jaca (Spain) Macroscopic electrical circuit

Resistor (R) Capacitor (C) Inductor (L)

Standard toolkit

✗

|0 |1 |2 |3

Energy

ωLC = 1 √ LC

Flux, !

| | | | V (t)

= V0 cos ω01t

Resistor (R) Capacitor (C) Inductor (L)

Standard toolkit

✗

• 19

Laura Cardani, INFN - Roma Low Radioactivity Techniques, Jaca (Spain) Quite simple to design and fabricate

Resistor (R) Capacitor (C) Inductor (L)

Standard toolkit

✗

|0 |1 |2 |3

Energy

ωLC = 1 √ LC

Flux, !

| | | | V (t)

= V0 cos ω01t

Resistor (R) Capacitor (C) Inductor (L)

Standard toolkit

✗

Josephson junctions

|0 |1 |2 |3

• Energy

Flux, !

V (t) = V0 cos ω01t

Very short π-pulse time

Credits for the figures to A. Blais, Université de Sherbrooke, Canada

BUILDING SUPERCONDUCTING QUBITS

• 20

THE NIGHTMARE OF QUASI- PARTICLES IN QUBITS

Laura Cardani, INFN - Roma Low Radioactivity Techniques, Jaca (Spain) Superconductor cooled to mK temperature —> electrons bound in Cooper pairs BUT Many mechanisms can break Cooper pairs, producing quasiparticles PROBLEM

• Absorb qubit energy and dissipate it via phonons
• Excite the qubit to |1> level
• Change the ΔE between |0> and |1>
• ….
• 21

PRODUCTION OF QUASI- PARTICLES IN QUBITS

Laura Cardani, INFN - Roma Low Radioactivity Techniques, Jaca (Spain) HOW

• Infrared radiation
• Thermal stress
• Dissipation of the readout line
• ….
• Phonons

What is the origin of quasiparticles bursts? How will they impact the performance of qubits?

• 22

QUBITS vs RESONATORS

Laura Cardani, INFN - Roma Low Radioactivity Techniques, Jaca (Spain)

• 23

140 keV 200 keV 280 keV

~ 4 µ/min with an E 150-300 keV

KID (Kinetic Inductance Detector) Developed within the CALDER project (ERC starting grant, PI: M. Vignati) In qubit, phonons are a nightmare On the contrary, in Particle Physics, phonons are what we want to measure!The technologies have opposite requirements but can benefit from each others

QUBITS vs RESONATORS

Laura Cardani, INFN - Roma Low Radioactivity Techniques, Jaca (Spain)

• 24

Entries 4927 Mean 21.02 RMS 35.47 Integral 4927

Energy [keV]

1 10

2

10

3

10 1 10

2

10

Entries 4927 Mean 21.02 RMS 35.47 Integral 4927

Entries 967 Mean 38.22 RMS 58.13 Integral 967 Entries 967 Mean 38.22 RMS 58.13 Integral 967

• ptical fiber

55Fe

Flat distribution from a few keV to ~300 keV (above detector not working) Rate (>10 keV) = 30 events/min detector