Nonlinear nanoelectromechanics of a movable Cooper-pair box

A. Patra; H. C. Park; L. Y. Gorelik; M. J. Park; R. I. Shekhter; S. Park

REVIEW 3 major objections 4 minor 21 references

Nonlinear nanoelectromechanics of a movable Cooper-pair box

T0 review · 3 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A movable Cooper-pair box loses mechanical stability at small eta and regains it at larger eta, with the nature of the second transition set by the tunneling rate relative to the mechanical frequency.

desk verdict Reentrant stability is real; the adiabatic/nonadiabatic bifurcation split is under-supported by a three-point table with unstated Q and epsilon. read the letter →

arxiv 2608.05012 v1 pith:K5QASF5C submitted 2026-08-05 cond-mat.supr-con

S. Park , A. Patra , L. Y. Gorelik , M. J. Park , H. C. Park , R. I. Shekhter This is my paper · ORCID

classification cond-mat.supr-con

keywords Cooper-pairboxnanoelectromechanicsAndreevtunnelingHopfbifurcationreentrantstabilityself-sustainedvibrationsnonadiabaticdynamicscentermanifoldnormalform

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper predicts that a movable Cooper-pair box coupled to a normal-metal pillar can switch between stillness and self-sustained vibration as a single control parameter, $\eta$ (the ratio of electrostatic energy to Josephson coupling energy), is increased. At small $\eta$ the resting configuration loses stability through a supercritical Hopf bifurcation and a stable vibration grows; at a larger $\eta$ the resting configuration regains stability through a second Hopf bifurcation. The authors show that the character of the second transition changes with adiabaticity: an inverse subcritical Hopf bifurcation when the Andreev tunneling rate far exceeds the mechanical frequency, and an inverse supercritical Hopf bifurcation when the two rates are comparable. These results extend earlier adiabatic self-vibration theory to the nonadiabatic regime and outline a nonlinear dynamical phase diagram for a superconducting electromechanical device.

What carries the argument

The argument is carried by the Hopf bifurcation normal form obtained from a center-manifold reduction of the coupled mechanical and Bloch-vector equations. On the two-dimensional center manifold the radial dynamics is $\dot r = \beta\,\zeta(\eta)\,r + a_1 r^3 + O(r^5)$, where $\zeta(\eta)=\eta-\eta_c$ and $\beta = d\operatorname{Re}\mu/d\eta$ at criticality; the sign of $a_1$ (the first Lyapunov coefficient) decides whether the bifurcation is super- or subcritical, while $\beta$ decides which side of $\eta_c$ the limit cycle lives on. Numerical continuation of the equilibrium branch supplies the eigenvalues and $a_1$ values collected in Table I, and those signs are what turn the spectral picture into the claimed subcritical/supercritical classification.

What would settle it

Numerically scan $a_1$ at $\eta_{c2}$ over a grid of nonadiabatic parameters with different quality factor $Q$ and coupling $\epsilon$; if the sign flips from negative to positive, the claimed inverse supercritical character is not robust. Experimentally, measure the vibration amplitude as $\eta$ is swept up and down through $\eta_{c2}$: a continuous drop to zero indicates supercritical, while a jump with hysteresis indicates subcritical.

Watch

Extended reading notes

Core claim

The central claim is that the device shows reentrant stability in $\eta$: the resting fixed point is stable below $\eta_{c1}\approx0.87$–$1.06$, becomes unstable in an interval, and is stable again above $\eta_{c2}\approx12$–$13.3$ for the parameter sets considered. The first instability is always a supercritical Hopf bifurcation, producing a stable small-amplitude limit cycle. The second transition, where stability is regained, is classified by the sign of the first Lyapunov coefficient $a_1$: in the adiabatic regime $E_J=\Gamma\gg\hbar\omega_0$ the coefficient is positive at $\eta_{c2}$, making the transition an inverse subcritical Hopf bifurcation, while in the nonadiabatic regime $E_J=\Gamma\sim\hbar\omega_0$ it is negative, making the transition an inverse supercritical Hopf bifurcation. The physical mechanism is competition between Andreev-tunneling pumping, which destabilizes the equilibrium, and field-induced Coulomb blockade of Cooper-pair tunneling, which suppresses the Josephson force as $\eta$ grows.

Load-bearing premise

The sign of the Lyapunov coefficient at the second critical point in the nonadiabatic regime is inferred from one computed parameter set and is close to zero; if it changes sign when damping or coupling changes, the claimed adiabatic-vs-nonadiabatic classification would not hold as stated.

Editorial extensions

If this is right

Self-sustained vibrations of the movable Cooper-pair box survive outside the adiabatic limit, so Cooper-pair shuttling is not restricted to slow mechanical motion.
Sweeping $\eta$ upward through $\eta_{c2}$ switches the vibration off either continuously (nonadiabatic inverse supercritical) or with bistability and a finite amplitude jump (adiabatic inverse subcritical).
In the nonadiabatic regime the two limit cycles coexist just below $\eta_{c2}$, so the same bias point can support small- and large-amplitude mechanical oscillations depending on initial conditions.
Above $\eta_{c2}$, field-induced Coulomb blockade suppresses Cooper-pair tunneling and the Josephson force, which is why the fixed point regains stability.
The $\eta$–$E_J/\hbar\omega_0$ plane contains a reentrant dynamical phase diagram whose boundaries are set by the two Hopf points.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

An experimental signature to look for: ramping the electric field up and down through $\eta_{c2}$ in the nonadiabatic regime should give a smooth amplitude collapse with little hysteresis, whereas the adiabatic regime should show a jump and hysteresis; this distinction is directly testable in the Andreev current noise or displacement signal.
Because $\eta$ is set by the external field, the device behaves as a field-controlled self-oscillator whose on/off threshold and hysteresis can be engineered, which may be useful for nanomechanical switching or sensing even though the paper does not pursue applications.
The same competition between pumping and Coulomb blockade may appear in other movable superconducting islands with position-dependent Josephson coupling, so similar reentrant stability windows are plausible in related shuttling devices.
The numerically observed smallness of $a_1$ at $\eta_{c2}$ in the nonadiabatic case suggests the supercritical/subcritical boundary could shift with damping or coupling strength, so a wider parameter scan is needed before treating the dichotomy as universal.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Quick take: this is a clean extension of the authors' earlier adiabatic analysis of a movable Cooper-pair box to the nonadiabatic regime. The reentrant stability in the control parameter eta is well supported and is the paper's main asset. The sharper claim—that the second Hopf bifurcation flips from inverse subcritical (adiabatic) to inverse supercritical (nonadiabatic)—is plausible but not adequately supported by the three-point table.

The paper does several things well. The model is clearly set out, the semiclassical reduction is standard, and the center-manifold appendix gives enough detail to follow the calculation. Using BifurcationKit with pseudo-arclength continuation is the right tool. The eigenvalue plots across the three parameter sets are consistent, and the reentrant stability part of the story holds in all of them. That is a genuine extension of Ref. [9], which only considered adiabatic motion.

The soft spots are real, and they are clustered around Table I. First, the quality factor Q and the coupling epsilon are never stated. They appear in Eqs. (8)-(10), and they control both the location of eta_c2 and the value of the first Lyapunov coefficient a1. Without them the entire numerical table is unreproducible. Second, the nonadiabatic regime is represented by a single point (EJ=Gamma=1.0), where a1=-0.0056 is small enough that its sign could easily change with Q or epsilon. Third, because a1 goes from +0.008 at EJ=10 to -0.0056 at EJ=1, a Bautin (codimension-2) point lies somewhere in that interval. A one-parameter slice cannot tell whether the sign change is truly governed by adiabaticity or by crossing the Bautin manifold at the chosen (Q,epsilon). Finally, all three points lie on the diagonal Gamma=EJ; the claim about the adiabatic/nonadiabatic dichotomy is therefore about a specific path, not about the full two-parameter space.

These are fixable problems. The authors should state Q and epsilon, and ideally provide a two-parameter scan (at least a1 as a function of EJ/Gamma at fixed Q and epsilon) to show the sign change is monotone and tied to adiabaticity. The reentrant stability result itself does not depend on that scan.

Who is this for: people working on nanoelectromechanical devices, superconducting shuttles, and nonlinear dynamics of open quantum systems. It deserves a serious referee, but the referee will need to ask for the missing parameters and a modest continuation before endorsing the central classification. I would cite the reentrant-stability result; I would not cite the bifurcation-type dichotomy in its present form.

Referee Report

3 major / 4 minor

Summary. The paper analyzes a semiclassical model of a movable Cooper-pair box coupled to a normal-metal pillar, with equations of motion given in Eqs. (10)-(11). Linearizing about the fixed point as a function of η, the ratio of electrostatic to Josephson energy, the authors find reentrant stability: the fixed point loses stability at a first Hopf bifurcation η_c1 and regains it at a second Hopf point η_c2. Center-manifold reduction and numerical continuation with BifurcationKit.jl are used to compute the first Lyapunov coefficient a1 at these points for three parameter sets (E_J=Γ=150, 10, 1). On this basis the paper claims that the second transition is inverse subcritical in the adiabatic regime and inverse supercritical in the nonadiabatic regime, and discusses the resulting limit-cycle coexistence.

Significance. The reentrant-stability result itself is consistently obtained in Fig. 1 and is not in question. If the adiabatic/nonadiabatic dichotomy for the second Hopf bifurcation is confirmed, the paper would extend the adiabatic self-vibration mechanism of Ref. [9] to a broader parameter regime and would give concrete predictions for limit-cycle coexistence in a superconducting nanoelectromechanical device. The use of standard numerical continuation with BifurcationKit.jl and an explicit center-manifold framework are strengths. However, the central classification rests on a small number of numerical points and on parameters that are not disclosed, so the significance is presently conditional.

major comments (3)

[Table I and Eqs. (8)-(11)] The numerical values of Q and ϵ are never stated, even though Q enters Eq. (11) through γ=Q^{-1} and ϵ is defined in Eq. (8) and multiplies the mechanical forcing terms. Both parameters affect the fixed point, the critical values η_c1 and η_c2, and the center-manifold coefficients that determine a1. The three rows of Table I are therefore not reproducible, and the sign of a1 at η_c2 could change with the unspecified Q and ϵ. Please report these values for every row and show that the bifurcation classification is robust under their variation.
[Table I, E_J=Γ=1.0] The nonadiabatic regime is represented by a single numerical point, with a1=-0.0056 at η_c2; the magnitude is close to zero and no numerical error estimate is given. Because a1 changes sign between E_J=Γ=10 (a1=0.0080) and E_J=Γ=1.0 (a1=-0.0056), a Bautin (codimension-2) point lies somewhere in that interval. Without a two-parameter continuation in (E_J, η), and without checking the dependence on Q and ϵ, the conclusion that the sign change follows the adiabatic/nonadiabatic crossover rather than the crossing of a Bautin manifold is not robustly established. The abstract's general claim about the nonadiabatic regime goes beyond what a single near-zero Lyapunov coefficient supports.
[Eq. (12a) and Table I] The distinction between an 'inverse' and a 'direct' Hopf bifurcation depends on the sign of β = d Re μ/dη at the critical point, since the existence condition for the small limit cycle is β δη/a1 < 0 (Eq. (14)). The manuscript never reports β or the product β δη at either Hopf point. Consequently, the labels 'inverse subcritical' and 'inverse supercritical' in Table I and the abstract are not directly supported by the tabulated quantities; please report β at η_c1 and η_c2 for all rows.

minor comments (4)

[Appendix A 1] The text says the continuation follows the '8D equilibrium branch', but Eq. (11) has seven state variables (x, v_x, y, v_y, m_1, m_2, m_3); please correct this to 7D.
[Section III] The term 'inverse' is used in the abstract and in Sec. IV but is not defined in the main text; please define 'inverse' in Sec. III where β is introduced.
[Table I caption] Please state the numerical tolerance and the normal-form convention used by BifurcationKit.jl for the Lyapunov coefficient computation, since a1 is convention-dependent.
[Sec. IV, last paragraph] The coexistence of two distinct limit cycles for η ≲ η_c2 in the nonadiabatic case is asserted without a figure or direct periodic-orbit computation; a representative phase portrait or limit-cycle continuation would support this statement.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the Hopf-bifurcation classification is computed from the stated model equations via center-manifold normal form, not fitted or assumed.

full rationale

The central claims, reentrant stability and the change in character of the second Hopf bifurcation, are obtained by solving the explicitly stated semiclassical equations of motion (Eq. 11) and computing the first Lyapunov coefficient a1 from the Poincare-Birkhoff normal form in Appendix A. The sign of a1 at eta_c2 is a numerical output of BifurcationKit.jl, not an input or a fitted parameter. References [9] and [16] are used for the model setup and for the standard center-manifold reduction, but the paper restates the model and the reduction steps, and the specific new result, the sign change of a1 between the adiabatic and nonadiabatic regimes, is not imported from those references. The concern that the nonadiabatic regime is represented by a single computed point with unstated Q and epsilon is a robustness or reproducibility issue, not circularity, because no step defines the claimed prediction in terms of the data used to make it. The paper is therefore self-contained against the stated dynamical equations and standard bifurcation theory.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The model relies on a set of standard approximations and unspecified numerical parameters; no new physical entities are introduced.

free parameters (4)

E_J (Josephson energy) = 150.0, 10.0, 1.0 (in units of hbar*omega0)
Chosen to sample adiabatic and nonadiabatic regimes; the sign of a1 at eta_c2 changes between E_J=10 and 1, so the classification is tied to these specific values.
Gamma (Andreev tunneling rate) = set equal to E_J
The Andreev tunneling rate is set equal to E_J in all runs; no scan over Gamma/E_J is presented, so the reentrant behavior may depend on this ratio.
epsilon (electromechanical coupling) = unspecified
The electromechanical coupling epsilon appears in the equations of motion and affects the fixed point and Jacobian, but its numerical value is not given in the results section.
Q (quality factor) = unspecified
The damping gamma = Q^{-1} enters the linear stability analysis; the value used in Fig. 1 and Table I is not stated.

assumptions (5)

domain assumption Born-Markov approximation
Used to derive the reduced Bloch equations (10b); requires the inequalities in Eq. (7).
domain assumption Semiclassical factorization of the density matrix
Treats displacement as classical and factorizes the density matrix; valid for lambda >> x0 and small epsilon, Eq. (8).
domain assumption Constant Andreev tunneling amplitude t_A
Assumed independent of mechanical displacement; simplifies Eq. (4).
domain assumption Superconducting parity effect with Coulomb blockade removed
The CPB is tuned to charge degeneracy via gate voltage; relies on Ref. [11].
standard math Center manifold and Hopf bifurcation theorems
Standard results used in the normal form analysis of Appendix A.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Nonlinear nanoelectromechanics of a movable Cooper-pair box." pith.science (2026). https://pith.science/paper/K5QASF5C

@misc{pith2026260805012,
  author       = {Pith},
  title        = {Pith review of: Nonlinear nanoelectromechanics of a movable Cooper-pair box},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/K5QASF5C}},
  note         = {Machine review of arXiv:2608.05012}
}

abstract

We theoretically study the dynamics of a movable Cooper-pair box coupled to a normal-metal pillar using a semiclassical approach. We analyze the dynamical stability induced by the nonlinear nanoelectromechanical coupling between the mechanical motion and an inelastic Andreev tunneling through linear stability and bifurcation analyses. As a function of $\eta$, defined as the ratio of electrostatic energy to Josephson coupling energy, the system exhibits reentrant stability. At small $\eta$, the fixed point loses stability through a supercritical Hopf bifurcation, giving rise to self-sustained vibrations. With a further increase of $\eta$, a second critical point appears, at which the fixed point regains stability. We show that this second transition corresponds to an inverse subcritical Hopf bifurcation in the adiabatic regime and to an inverse supercritical Hopf bifurcation in the nonadiabatic regime. These results extend previous studies of adiabatic self-vibrations to the nonadiabatic regime and reveal a rich nonlinear dynamical phase diagram arising from the interplay between electronic and mechanical degrees of freedom in superconducting devices.

Figures

Figures reproduced from arXiv: 2608.05012 by the authors.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

21 extracted references · 20 canonical work pages

[9]

Paraﬁlo, L

A.V. Paraﬁlo, L. Y. Gorelik, M. V. Fistul, H.C. Park, and R. I. Shekhter, Nanomechanics driven by Andreev tunneling, Phys. Rev. B 102, 235402 (2020)

work page 2020
[1]

The continued equilibrium branch is parameterized as ( X0(s), η(s)), where s is an auxiliary continuation parameter corresponding to the pseudo-arclength along the branch

Continuation of Equilibria Using BifurcationKit.jl Additionally, we must clarify that the nu- merical bifurcation analysis is carried out using BifurcationKit.jl, which follows the 8D equilibrium branch by Pseudo-Arclength Continuation (PALC). The continued equilibrium branch is parameterized as ( X0(s), η(s)), where s is an auxiliary continuation paramet...

work page
[2]

= hi1 (X ′ 1)2 + hi2 (X ′ 2)2 + hi3X ′ 1X ′ 2 + higher-order terms, i = 3, . . . ,7. (A8) These real coeﬃcients hij, for i = 3, . . . ,7 and j = 1, 2, 3 will be obtained using Eq. (A10). In the coordinates introduced using Eqs. (A4), (A5), (A6), and (A7), the CPB dynamics (A1) takes the form ( ˙X ′ 1 ˙X ′ 2 ) = ( ζ(ξ) ω(ξ) −ω(ξ) ζ(ξ) )    Ac ( X ′ 1 X...

work page
[3]

Substituting the resulting approximation for hc (X ′ c) into Eq

These coef- ﬁcients are suﬃcient to construct the Poincar´ e-Birkhoﬀ normal form to cubic order and thereby determine the nature of the Hopf bifurcation. Substituting the resulting approximation for hc (X ′ c) into Eq. (A9a), we obtain the ﬂow on the center manifold. The next step is to eliminate nonessential nonlinearities through a sequence of near-iden...

work page
[4]

A. N. Cleland, Foundations of Nanomechanics (Springer, New York, 2003)

work page 2003
[5]

LaHaye, J

M. LaHaye, J. Suh, P. Echternach, K. C. Schwab, and M. L. Roukes, Nanomechanical measurements of a su- perconducting qubit, Nature 459, 960 (2009)

work page 2009
[6]

Pirkkalainen, S

J.-M. Pirkkalainen, S. U. Cho, J. Li, G. S. Paraoanu, P. J. Hakonen, M. A. Sillanp¨ a¨ a,Hybrid circuit cavity quan- tum electrodynamics with a micromechanical resonator , Nature 494, 211 (2013)

work page 2013
[7]

L. Y. Gorelik, A. Isacsson, Y. M. Galperin, R. I. Shekhter, and M. Jonson, Coherent transfer of Cooper 8 pairs by a movable grain , Nature 411, 454 (2001)

work page 2001

Show all 21 references

[8]

L. Y. Gorelik, A. Isacsson, M. V. Voinova, B. Kasemo, R. I. Shekhter, and M. Jonson, Shuttle mechanism for charge transfer in Coulomb blockade nanostructures , Phys. Rev. Lett. 80, 4526 (1998)

1998
[10]

Fedorets, L

D. Fedorets, L. Y. Gorelik, R. I. Shekhter, and M. Jonson, Spintronics of a nanoelectron mechanical shuttle , Phys. Rev. Lett. 95, 057203 (2005)

2005
[11]

S. Park, L. Y. Gorelik, S. I. Kulinich, H. C. Park, C. Kim, R. I. Shekhter, Andreev probing of a Cooper-pair ﬂying qubit, Phys. Rev. B 111, 165403 (2025)

2025
[12]

S. Park, A. Paraﬁlo, L. Y. Gorelik, and R. I. Shekhter, Adiabatic self-vibrations of a movable Cooper- pair box generated by inelastic Andreev tunneling, arXiv:2604.10041 (2026)

2026 arXiv
[13]

J. D. Crawford, Introduction to bifurcation theory , Rev. Mod. Phys. 63, 991 (1991)

1991
[14]

K. A. Matveev, M. Gisself¨ alt, L. I. Glazman, M. Jon- son, R. I. Shekhter, Parity-induced suppression of the Coulomb blockade of Josephson tunneling , Phys. Rev. Lett. 70, 2940 (1993)

1993
[15]

L. Y. Gorelik, D. Fedorets, R. I. Shekhter, and M. Jonson, Spin-controlled nanoelectromechanics in magnetic NEM- SET systems , New J. Phys. 7, 242 (2005)

2005
[16]

Breuer and F

H.-P. Breuer and F. Petruccione, The Theory of Open Quantum Systems (Oxford University Press, New York, 2002)

2002
[17]

Y. A. Kuznetsov, Elements of Applied Bifurcation The- ory, (Springer-Verlag, New York, 1998)

1998
[18]

R. C. Hilborn, Chaos and Nonlinear Dynamics, An Intro- duction for Scientists and Engineers , (Oxford University Press, New York, 2001)

2001
[19]

Patra, B

A. Patra, B. L. Altshuler, and E. A. Yuzbashyan, Driven- Dissipative Dynamics of Atomic Ensembles in a Reso- nant Cavity: Nonequilibrium Phase Diagram and Pe- riodically Modulated Superradiance, Phys. Rev. A 99, 033802 (2019)

2019
[20]

(B6) of ref

An example of such linear transformation is given in Eq. (B6) of ref. [16]
[21]

Veltz, BifurcationKit.jl: bifurcation analysis in Julia , HAL preprint hal-02902346 (2020)

R. Veltz, BifurcationKit.jl: bifurcation analysis in Julia , HAL preprint hal-02902346 (2020)

2020

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [9]

Paraﬁlo, L

A.V. Paraﬁlo, L. Y. Gorelik, M. V. Fistul, H.C. Park, and R. I. Shekhter, Nanomechanics driven by Andreev tunneling, Phys. Rev. B 102, 235402 (2020)

work page 2020

[2] [1]

The continued equilibrium branch is parameterized as ( X0(s), η(s)), where s is an auxiliary continuation parameter corresponding to the pseudo-arclength along the branch

Continuation of Equilibria Using BifurcationKit.jl Additionally, we must clarify that the nu- merical bifurcation analysis is carried out using BifurcationKit.jl, which follows the 8D equilibrium branch by Pseudo-Arclength Continuation (PALC). The continued equilibrium branch is parameterized as ( X0(s), η(s)), where s is an auxiliary continuation paramet...

work page

[3] [2]

= hi1 (X ′ 1)2 + hi2 (X ′ 2)2 + hi3X ′ 1X ′ 2 + higher-order terms, i = 3, . . . ,7. (A8) These real coeﬃcients hij, for i = 3, . . . ,7 and j = 1, 2, 3 will be obtained using Eq. (A10). In the coordinates introduced using Eqs. (A4), (A5), (A6), and (A7), the CPB dynamics (A1) takes the form ( ˙X ′ 1 ˙X ′ 2 ) = ( ζ(ξ) ω(ξ) −ω(ξ) ζ(ξ) )    Ac ( X ′ 1 X...

work page

[4] [3]

Substituting the resulting approximation for hc (X ′ c) into Eq

These coef- ﬁcients are suﬃcient to construct the Poincar´ e-Birkhoﬀ normal form to cubic order and thereby determine the nature of the Hopf bifurcation. Substituting the resulting approximation for hc (X ′ c) into Eq. (A9a), we obtain the ﬂow on the center manifold. The next step is to eliminate nonessential nonlinearities through a sequence of near-iden...

work page

[5] [4]

A. N. Cleland, Foundations of Nanomechanics (Springer, New York, 2003)

work page 2003

[6] [5]

LaHaye, J

M. LaHaye, J. Suh, P. Echternach, K. C. Schwab, and M. L. Roukes, Nanomechanical measurements of a su- perconducting qubit, Nature 459, 960 (2009)

work page 2009

[7] [6]

Pirkkalainen, S

J.-M. Pirkkalainen, S. U. Cho, J. Li, G. S. Paraoanu, P. J. Hakonen, M. A. Sillanp¨ a¨ a,Hybrid circuit cavity quan- tum electrodynamics with a micromechanical resonator , Nature 494, 211 (2013)

work page 2013

[8] [7]

L. Y. Gorelik, A. Isacsson, Y. M. Galperin, R. I. Shekhter, and M. Jonson, Coherent transfer of Cooper 8 pairs by a movable grain , Nature 411, 454 (2001)

work page 2001