• Aucun résultat trouvé

THERMOELECTRIC BRANCH IMBALANCE CURRENTS IN SUPERCONDUCTORS

N/A
N/A
Protected

Academic year: 2021

Partager "THERMOELECTRIC BRANCH IMBALANCE CURRENTS IN SUPERCONDUCTORS"

Copied!
3
0
0

Texte intégral

(1)

HAL Id: jpa-00217639

https://hal.archives-ouvertes.fr/jpa-00217639

Submitted on 1 Jan 1978

HAL is a multi-disciplinary open access

archive for the deposit and dissemination of

sci-entific research documents, whether they are

pub-lished or not. The documents may come from

teaching and research institutions in France or

abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est

destinée au dépôt et à la diffusion de documents

scientifiques de niveau recherche, publiés ou non,

émanant des établissements d’enseignement et de

recherche français ou étrangers, des laboratoires

publics ou privés.

THERMOELECTRIC BRANCH IMBALANCE

CURRENTS IN SUPERCONDUCTORS

D. Heidel, J. Garland

To cite this version:

(2)

JOURNAL DE PHYSIQUE Colloque C6, supplt!ment au no 8, Tome

39, aofit

1978, page C6-492

THEWOELECTRIC BRANCH IMBALANCE CURRENTS I N SUPERCONDUCTORS

D.F. Heidel and J.C. Garland

Physics Deparhent, The Ohio S t a t e University CoZwnbus, Ohio, U.S.A. 43210

REsum6.- Un gradient de tempsrature appliqu6 B un supraconducteur produitun courant d'excitations de quasiparticules accompagns par un contre-courant superfluide. Les solutions des Equations cinE- tiques pour les quasiparticules en prssence de ce courant superfluide montrent l'existence d'un Etat hors dlEquilibre dans le spectre d'excitation des quasiparticules. Ce dEsEquilibre est de l'ordre de v et donne un terme d'ordre vs2 au courant de quasiparticules.

Abstract.- A temperature gradient applied to a superconductor produces a current of quasiparticle excitationswhich is accompanied by a counter-flow of superfluid current. Solutions of the quasipar- ticle kinetic equation in the pr6sence of this superfluid current show the existence of a branch imbalance in the quasiparticle excitation spectrum which is of order v and which contributes to the quasiparticle current a term of order v

-

A temperature gradient applied to a bulk iso- G =

k

-

FkO tropic superconductor will produce a current of now

where

equilibrium quasiparticle excitations

5.

Because

-

-

Fko = [exp(8Ek) +

g-'

( 6 )

Sn

is always accompanied by a counterflow of super-

4

current it is important to allow for the influ- FkO should be distinguished from the equilibrium

S'

ence of condensate motion on the quasiparticle dis- distribution F k given by

tribution function. We have investigated the effects FkO = Lexp(~'k) +

Q-'

(7)

of a uniform superfluid flow on the solution of the quasiparticle kinetic equation. Our primary finding is that an imbalance between the electron-like exci- tation spectrum is induced by the superfluid flow. This branch imbalance /l /, denoted by Q ~ , is first order in vs and contributes a term to the normal current of order vs2. This branch current may beco- me important near the superconducting transition

temperature where vs becomes divergent 121. In the presence of a uniform superfluid flow the excitation spectrum is given by 131.

In the presence of a temperature gradient Equation

where the collision operator has been separated into a quasiparticle impurity operator and a qua- siparticle-phonon operator.

The expression for the total current density

+

J (i.e., that carried by the quasiparticles and by the condensate) is + - + + J = e

1

[Uk2~k + vk2 (1

-

F-~)J (vk + vs) (9) k

-

+ -f Ek = Ek + Mk*vs, (1) where Uk2 = 1

where Ek is the equilibrium excitation energy /2(1 +Sk/Ek) ; vk2 ='/~(I-E~/E~). (10) Ek =

kk2

+ n q 1 / 2

and where

5,

= ($k12/2m

-

1.1

(2) Using Equations (5) and (IO), Equation (9) can be written

-+ -+ aFko

- + +

-+

E,

(3) J=nevteI k (-1 aEk

(6

k .vs)vk+e GkGk+GS 1e(--)Gk Ek (I I) The steady state kinetic equation for the quasipar- The first two terms of Equation (1 1) correspond to ticle distribution function Fk with allowance for the superfluid current Ss, while the last two

+

the motion of the condensate

-

-

/4/ is terms represent the thermoelectric current Jn. The gEk

.

VrFk

-

VrEk

.

%Fk +

I C F ~ I

= 0 ( 4 ) third term in Equation (11) resembles a normal me- where I{F } is the operator that describes the col- tal thermoelectric current. In a superconductor,

k

lisions of the quasiparticles with impurities and however, this current must be calculated by allo- + phonons. A non-equilibrium distribution function wing for the dependence of Gk on vs. The fourth

X+

Gk can be defined as term in Equation (11) can be written as Q vs where

(3)

Q X , the charge-weighted branch imbalance, is defi- ned by /l/

Q= = e

1

(Sk/Ek)Gk. (1 2) k

This term produces a current whenever a branch im- balance exist in the superconductor.

-+

If v is assumed to be small enough to satis- P v

F S

fy the condition (p)l , <<the left hand side of

A -+

Equation (8) can be expanded in powers of vs. The solution to the zeroth order term,previously stu- died by GalTperin et a1.,/5/ can be completely relaxed by elastic scattering.The symmetries of

-+

this solution with respect to k and

Sk

can be seen in figure la), where the distribution function is plotted along the direction defined by the tempera- ture gradient. Gal1perin et al.have shown for T near Tc, the quasiparticle current due tothis dis- tribution is approximately equal to the thermoelec- tric current in the normal state at T = T

.

Fig.l : Plot of Gk along axis defined by the tempe- rature gradient : (a) zeroth order solution,(b) G1 s (C) G2.

The solution to Equation (8) that is of order v 2 has been studied by Aronov 161, who has shown that the resulting current cannot be relaxed by impurity scattering. Altohough the current propor- tional to v is smaller than that produced by the zeroth order solution by a factor (v P /A)~(T S F ph /

Timp), this current can still be important since

Where

(14)

In Equation (15),r is the branch imbalance relaxa-

Q

tion time. The symmetries of these components of G are shown in figure Ib) and lc).G1 represent a

k

net decrease in the total number of excitations, equivalent to a small increase in the magnitude of the energy gap 161. G2 produces the branch imbalan- ce current. A numerical integration of G2shows that G represents a net excess in the number of elec-

2

tron-like excitations over the number of hole-like excitations. The ratio of the branch imbalance cur- rent to the zeroth order current is approximately

V s P ~ 'Q

( T ) 2

C

.

Since rQ and v are both divergent

imp

near the transition temperature /l/, this current can become important for T": T

C'

References

/l/ Tinkham,M,; Phys. Rev. B

6

(1972) 1747. /2/ Garland,J.C., and Van Harlingen,D.J.,Phys.Lett.

47A (1974) 423.

131 de Gennes,P.G.,Superconductivity of Metals and Alloys, (W.A.Benjamin,New York, 1965)

/4/Ga11perin,Yu M.,Gurevich,V.L., and Kozub,V.I., Zh. Fksp Teor .Fiz 65 (1 973) 1045, (Sov.Phys .-JETP

-

38 ( 1 9 7 4 ) w ) F

-

-

The solutions to the first order term in Equation (8) in the relaxation time approximation are

= Gl + G2

Références

Documents relatifs

In this case the two positive integers |y| and |y − 1| are consecutive, therefore they are relatively prime. If the product of two relatively prime integers is a square, then each

To further exclude that EBP-2 contributes to creating the force imbalance, we analyzed the dynamics of dynein at the cortex as we did in the LSP, finding equal

While the losing order parameter disappears in the mean field ground state, it should manifest itself as a soft mode in the excitation spectrum.. Think for instance of a

The aim of this research is dual: first, to analyze different machine learning algorithms to classify text data and report the best performing one, and,

The proof of theorem 4.1 actually shows that the heat kernel associated to the Lichnerowicz operator is positivity improving in case (M n , g 0 (t)) t≥0 is an expanding gradient

[r]

In this paper, our first goal is to show that indeed there is a sharp estimate for λ 1 (M) under Ricci curvature lower bound. To prove this, we will develop a new argument, that

The results show that CNG with the proposed distance measures is more accurate when only limited training text samples are available, at least for some of the candidate authors,