The electrostatic potential and the associated charge distribution in the vortices of high-Tc superconductors involving mixed symmetry state of the order parameters have been studied. The work is carried out in the framework of an extended Ginzburg-Landau (GL) theory involving the Gorter-Casimir two-fluid model and Bardeen's extension of GL theory applied to the high-Tc superconductors. The properties are calculated using the material parameters relevant for the high-Tc cuprate YBCO.
1. Introduction
A key feature which characterizes the vortices in any type-II superconductor is the ability of the vortex to support a magnetic flux, with the magnetic flux quantum being defined as Φ0=hc/2e. However, what is less known and came into light only recently [1–3] is the fact that the vortices, along with supporting a magnetic flux quanta, can support an accumulation of finite electric charge in it. The accumulation of charge in the vortices is an artifact of the difference between the chemical potential inside the vortex core and of the region outside the core. The presence of diamagnetic electric current in any superconductor gives rise to inhomogeneity, as a result of which there arises an internal electric field in the superconductor, which maintain the neutrality of the charge distribution in the system and thus a constant electrochemical potential in the superconductor. The presence of an internal electric field in a superconductor with a stationary current was first discussed by Bopp [4] and such an electric field was later measured by a number of experiments [5, 6]. A uniform current in a superconductor results in an electric field analogous to the Bernoulli pressure variation associated with the nonuniform flow of a classical fluid [7]. The corresponding electric potential is thus termed as the Bernoulli potential [8] and is profoundly influenced by the band structure of the superconducting material [9, 10]. The experimentally observed Hall anomaly in high-temperature superconductors has been attributed to the charge accumulation in the vortex core. In case of the high-Tc cuprates, it has been observed that there is a sign reversal of the flux flow Hall coefficient below Tc [11]. It has been suggested that this Hall anomaly of high-temperature superconducting cuprates is universal and is dependent on the doping of the material, with a reversal of sign of the Hall coefficient from the overdoped to the underdoped regime. Kumagai et al. [12] have for the first time experimentally studied the accumulation of charge in the vortex cores in high-Tc superconductor by using high-resolution measurements of the nuclear quadrupole frequency which is sensitive to the local charge density. The behavior of the vortex core and the charge accumulation in the core in high-Tc cuprate are different from that of the conventional type-II superconductors owing to the complicated Fermi surface architecture in these materials. A theoretical model is thus required which can be used to study the electrostatic potential and the associated charge distribution in the votices of the high-Tc superconducting cuprates. In the present work, we attempt to do so in the framework of a phenomenological model. Such an approach has been used to study the electrostatic potential in conventional type-II superconductors [13, 14], but for the high-Tc superconducting cuprates such a study is still lacking.
For developing the theoretical model, it is important to discuss the pairing state symmetry of the high-Tc superconducting cuprates. In spite of the conflicting evidence regarding the pairing state symmetry of high-temperature superconducting cuprates over the past decades, a consensus could now be reached that even in case of a strictly tetragonal system [15, 16], the high-temperature superconducting cuprates possess a mixed symmetry state of the order parameter components (for details see [17]). The mixed pairing state symmetry is characterized by the presence of a dominant d-wave order parameter component along with a subdominant s-wave order parameter component [18–20]. It has been found that the properties of such system can be theoretically explained by allowing for two or more order parameter components and their derivative mixing terms in the free energy density functional [17, 21–25]. In the present work, a similar two-order parameter Ginzburg-Landau (GL) theory involving mixed symmetry state of the order parameter components is used to study the electrostatic properties of the high-Tc superconducting cuprate. The other important aspect of the present work is the extension of the GL theory to the low temperature regime. The GL theory is known to be applicable to the temperature regions near the critical temperature (Tc); however, it has been observed that even at temperatures much below the Tc (i.e., T≈(2/3)Tc) the GL theory provides very good result. At temperature lower than this, inaccuracies tend to set in the result obtained by employing the GL model. Extension of the present two-order parameter GL theory to the low temperature regime has been carried out in a manner parallel to the Bardeen's extension of the conventional GL theory [13, 14, 26]. The purpose of this paper is thus twofold, firstly it gives an extension of the two-order parameter GL theory corresponding to the high-Tc superconducting cuprates involving mixed symmetry state of the order parameter components to the low temperature regime, secondly it discusses about the electrostatic potential and charge distribution in the high-Tc superconducting system.
The rest of the paper is organized as follows: in Section 2 the theoretical formulation of the problem is discussed; Section 3 deals with the results obtained regarding the magnetic properties of the system and their analysis, while in Section 4 the details regarding the electrostatic properties have been discussed. Finally we conclude in Section 5 with the suggestion for future works.
2. Extended Ginzburg-Landau Model for High-Tc Cuprates
A second-order transition such as the normal-superconductor phase transition can be described by Gorter-Casimir two-fluid model [27]. In case of high-temperature superconductors involving mixed symmetry state of the order parameters, the d- and s-wave order parameter components can be expressed in terms of superconducting fractions ω¯d and ω¯s, respectively. In presence of the superconducting electrons the normal-state free energy density is modified asfs=〈U-ϵcondω¯d+ϵconsω¯s-12γT21+ω¯s12γT21+ω¯s-12γT21-ω¯d-12βsdγT21+ω¯s1-ω¯d〉12γT21+ω¯s.
The superconducting state is an ordered state. The transition of the electrons from the disordered normal state to the ordered superconducting state will give out a certain condensation energy which is expressed in (1) as ϵcondω¯d and ϵconsω¯s for the d- and s-wave type superconducting electrons, respectively. As can be seen from (1), in the present study the condition of stable d-wave configuration in the bulk has been considered, with a single transition temperature Td. The d-wave pairing interaction is considered to be attractive, while for the s-wave a repulsive interaction is considered [28, 29]. In (1) U is the internal energy of the system, while γ is the linear coefficient of specific heat. The terms (1/2)γT21+ω¯s and (1/2)γT21-ω¯d correspond to the reduction in the entropy of the system due to the ordering of the electrons in the superconducting state. The term (1/2)βsdγT21+ω¯s1-ω¯d corresponds to the contribution arising due to the interaction between the s- and d-wave order parameter components. The coefficient βsd has been kept as a variable so as to understand the effect of the interaction term. At the critical temperature Td, the ordering of the superconducting electron vanishes and the system return to the normal state, thus we can writeϵcon=ϵcons=ϵcond=14γTd2.
In terms of the total electron density n, the superconducting fractions are defined as ω¯s=2|s|2/n and ω¯d=2|d|2/n, respectively, with n=(2(|s|2+|d|2)+nn), where nn is the normal state electron density. Equation (1) gives the condensation energy density of the system.
We next write the contribution of the kinetic energy of the superconducting condensate towards the total free energy density of the system. In presence of the mixed symmetry state of the order parameter components, the kinetic energy density contribution can be written as
fkin=〈×[(Πys)*(Πyd)-(Πxs)*(Πxd)+c.c]][γs|Πs|2+γd|Πd|2+γv[(Πys)*(Πyd)-(ΠxsΠyd)*(ΠxdΠyd)+c.c]]〉,
where Π=(-i∇-2eA/ℏc) and γi is related to the effective electronic masses as γi=ℏ2/2mi* with i=s,d, and v. The above expression consists of three terms. The first term with the coefficient γs corresponds to the contribution of the s-wave order parameter component to the kinetic energy of the system, the second term with the coefficient γd gives the contribution of the d-wave order parameter component. Finally, the third term with the coefficient γv is the mixed gradient coupling term and combines the gradient contributions of the d- and s-wave type order parameter components. The mixed gradient coupling term is the most important term regarding the generation of s-wave order parameter component in the system. Previous theoretical studies of high-Tc superconductors carried out in the framework of two-order parameter GL theory have shown that properties of these materials are significantly affected by the admixture of s-wave order parameter component and the decisive role regarding the contribution of the s-wave order parameter component in the system is played by the mixed gradient coupling term.
It has been observed through the linear stability analysis that with the vanishing of the mixed gradient coupling term, the bulk d-wave solution is stable against the admixture of the s-wave order parameter component [16]. Thus, it can be said that in case of the standard two-order parameter GL model, the higher order coupling terms, namely |s|2|d|2 does not give rise to any significant contribution towards the admixture of the s-wave order parameter component. It must be noted that in case of the extended GL theory the model deals with temperatures much below the critical temperature where the density of the superconducting electrons gets enhanced. In order to verify the relative contribution of the d- and s-wave type superconducting electrons, the average value of the order parameter components has been calculated at different temperatures and magnetic field inductions. Irrespective of presence or absence of the interaction term with coefficient βsd, the magnitude of the s-wave order parameter component is found to be significantly smaller than that of the d-wave order parameter component. Even at a low temperature of t=0.5 and very low magnetic field, it has been observed that the relative magnitude of the d- and s-wave contribution does not change significantly with the presence or absence of the coupling term. This indicates the fact that even at low temperatures the principal contribution towards the s-wave order parameter component arises from the mixed gradient coupling term only. The observation is presented in Figure 1 for a significant value of the coupling term βsd=0.5, and it can be seen that over the wide range of magnetic field induction and specially at lower magnetic fields (where the superconducting electron density is enhanced), the relative magnitude of the d- and s-wave order parameter components remains unchanged for the presence and absence of the interaction term. The observation justifies the approximation that for studying the properties of the high-Tc superconductors in the framework of extended two-order parameter GL theory, it is sufficient to consider the contribution of the s-wave order parameter component arising from the mixed gradient coupling term only and thus for the remaining part of the present study one can set βsd=0.
Variation of relative magnitude of d- and s-wave order parameter components 〈ωd〉/〈ωs〉 with the coupling parameter βsd at different magnetic field inductions b. Other relevant parameters are t=T/Td=0.5, κ=72, γv/γd=ϵv=0.1.
Next the contribution of the electrostatic potential towards the free energy density of the system is discussed. The corresponding Coulomb energy as a part of the free energy density is expressed asfc=〈ϕρ-12ϵ0|∇ϕ|2〉,
where the total charge density is defined as ρ=en+ρlattice. In this expression, ρlattice represents the lattice charge density. The electrostatic potential is determined from the Coulomb interaction and can be given asϕ(r)=∫dr′14πϵ1|r-r′|ρ(r′)
or in its differential form by the Poisson equation-ϵ∇2ϕ=ρ.
Finally, we take into account the contribution of the magnetic energy towards the free energy density of the system. For the applied magnetic field Ba the resulting magnetic energy density is given asfM=〈12μ0(B-Ba)2〉=〈18π(B-Ba)2〉,
where μ0=4π. Using (1)–(7), the resulting free energy density can be expressed asf=〈U+14γTd22|s|2n-14γTd22|d|2n-12γT21+2|s|2n+γs|Πs|2+γd|Πd|2-12γT21-2|d|2n-12γT2βsd1+2|s|2n1-2|d|2n+γv[(Πys)*(Πyd)-(ΠxsΠyd)*(ΠxdΠyd)+c.c]+ϕ(en+ρlattice)-12ϵ0|∇ϕ|2+18π(B-Ba)212γT21+2|s|2n〉,
with n=(2|s|2+2|d|2+nn) being the total electron density. Equation (8) gives the extended two-order parameter Ginzburg-Landau (GL) free energy density for the high-Tc superconductors involving mixed symmetry state of the order parameters. Near the critical temperature (Td) the equation reduces to the standard two-order parameter Ginzburg-Landau free energy density functional, consisting of higher-order interaction terms of the order parameter components. The total free energy is a function of the order parameter components s and d, the vector potential A, and the normal state electron density nn. The other physical quantities, namely, B, ϕ, n, ρ, and so forth are to be understood in terms of s, d, A, and nn. The variables are dependent on the material parameter U, Td, γ, ms*=md*, ρlattice, ϵ0, and n. Since the density dependence of the material parameters lead to significant corrections, we assume that U, Td, and γ are dependent on the density n. However, ϵ0, ms*=md*, ρlattice, and so forth are taken as constants.
We now proceed to determine the equations of motion from the extended GL-free energy density functional. On minimizing (8) with respect to the electrostatic potential ϕ, the order parameters (s and d), the vector potential A, and the normal state electron density nn, we get a complete set of equations. Equations (9)–(13) comprising of the Poisson equation, GL equations corresponding to the d- and s-wave order parameter components, Ampere's law and Bernoulli potential, respectively, give the complete set of equations corresponding to the high-Tc superconductors involving mixed symmetry state of order parameters in presence of electrostatic potential,
-ϵ0∇2ϕ=e(nn+2|s|2+2|d|2)+ρlattice,0=γdΠ2d+γv(Πy2-Πx2)s+(1+2|s|2/n1-2|d|2/n-γTd22n+11-2|d|2/nγT22n+γT22nβsd×1+2|s|2/n1-2|d|2/n)d,0=γsΠ2s+γv(Πy2-Πx2)d+(γT22nβsd1-2|d|2/n1+2|s|2/nγTd22n-11+2|s|2/nγT22n-γT22nβsd1-2|d|2/n1+2|s|2/n)s,∇2A=4π{(-ℏe*ms*)[s*(Πs)+s(Πs)*]+(-ℏe*md*)[d*(Πd)+d(Πd)*]+(-ℏe*mv*)(ŷ[s*(Πyd)+d(Πys)*+c.c]-x̂[s*(Πxd)+d(Πxs)*+c.c]d(Πys)*)(-ℏe*ms*)},eϕ=λTF2∇2eϕ-(1n){γdd*Π2d+γvd*(Πy2-Πx2)s}-(1n){γss*Π2s+γvs*(Πy2-Πx2)d}-(|s|22n)Td2(∂γ∂n)+(|d|22n)Td2(∂γ∂n)-(|s|2n)γTd(∂Td∂n)+(|d|2n)γTd(∂Td∂n)+(T22)(∂γ∂n)(1+2|s|2n)+(T22)(∂γ∂n)(1-2|d|2n)+(T22)(∂γ∂n)×βsd1+2|s|2n1-2|d|2n,
where the contribution λTF2∇2eϕ arises due to the Thomas-Fermi screening. In the above equations, we have used the effective potential χ=χd+χs acting on the s- and d-wave type electrons. The effective potentials corresponding to d-and s-wave type electrons are defined as
χd=-2nϵcon+γT22n11-2|d|2/n+γT22nβsd1+2|s|2/n1-2|d|2/n,χs=2nϵcon-γT22n11+2|s|2/n-γT22nβsd1-2|d|2/n1+2|s|2/n.
In the above set of equations, the basic material parameters are γ, Td, ms*=md*, ∂Td/∂n, and ∂γ/∂n.
3. Magnetic Properties of the System
We begin the discussion by studying the magnetic properties of high-temperature superconductors involving mixed symmetry state of the order parameters. In terms of reduced unit, the quantities used in the calculation are expressed as
t=TTd,B′=λLonBλ0BcB0,A′=Aλ0BcB0,r′=rλLon,λLon=λ01-t4,κ(t)=κ021+t2,Bc(t)=B0(1-t2),Bc2(t)=2κBc,
where B0=Tdμ0γ/2, λ0=md/e2nμ0, and κ0=(mdTd/neℏ)γ/μ0. The extended GL free energy density functional is then expressed in terms of the gauge invariant real quantities ωd=2|d|2/n(1-t4), ωs=2|s|2/n(1-t4), and Q, where s(x′,y′)=ωs(x′,y′)eiϕs(x′,y′) and d(x′,y′)=ωd(x′,y′)eiϕd(x′,y′) correspond to the order parameter components, while Q′(x′,y′)=A′(x′,y′)-∇′ϕ(x′,y′)/κ is the velocity of the superconducting electrons. The corresponding two-dimensional free energy density in terms of dimensionless unit can be thus written asf′=〈-(-ωs+ωd)(1-t2)-2t21+ωs(1-t4)(1-t2)(1-t4)-2t2×1-ωd(1-t4)(1-t2)(1-t4)-2t2βsd1+ωs(1-t4)×1-ωd(1-t4)(1-t2)(1-t4)+gs+ωsQ′2+gd+ωdQ′2+ϵv[2cos(ϕ){(∇y′ωs)(∇y′ωd)4κ2(ωsωd)1/2-(∇x′ωs)(∇x′ωd)4κ2(ωsωd)1/2+(Qy′2-Qx′2)(ωdωs)1/214κ2(ωsωd)1/2}+2sin(ϕ){Qy′(∇y′ωs)2κ-Qx′(∇x′ωs)2κ}(ωdωs)1/2-2sin(ϕ){Qy′(∇y′ωd)2κ-Qx′(∇x′ωd)2κ}×(ωsωd)1/2]+(∇′×Q′-Ba′)22t21+ωs(1-t4)(1-t2)(1-t4)〉′,
where Ba′ is the applied magnetic field and f′=f/((1/4)γTd2(1-t2)(1-t4)). For studying the magnetic properties of the system, the contribution of the Coulomb energy and the internal energy to the free energy density functional is neglected. In the above equation, ϵv gives the strength of the s-wave order parameter component in the system and is defined as ϵv=γv/γd. We now discuss about the selection of the parameters in this work. Theoretical studies carried out on high-Tc superconductors involving mixed pairing state symmetry of order parameters suggested the value of mixed gradient coupling parameter to be ϵv=γv/γd≈0.1-0.4 [30]. Further, it was also observed that with ϵv=0.1 the theoretical results obtained in the framework of two-order parameter GL theory were in excellent agreement with the experimental data corresponding to YBa2Cu3O7 − δ [17, 22, 24, 25]. Thus, for the present study the mixed gradient coupling parameter ϵv has been chosen to be ϵv=0.1. The effects of the higher values of the coupling parameter ϵv have also been verified, so as to understand the influence of the admixture of s-wave order parameter component in the system at various temperatures. The results obtained for different values of ϵv have been found to be qualitatively the same, with difference in magnitude depending upon the amount of admixture of the s-wave order parameter component in the system. The effects of the admixture of s-wave order parameter component on the various properties of the high-Tc cuprates, namely, the vortex lattice structure, local spatial behavior of the order parameter and magnetic field profiles, reversible magnetization of the system, and shear modulus of the flux line lattice and so forth, studied in the framework of the standard two-order parameter GL model, have already been reported in the literature [17, 22, 24, 25].
The three GL equations obtained by minimizing the free energy density functional equation (16) with respect to the order parameter components ωs, ωd and supervelocity Q′ as, δf′(r′)/δωs(r′) = δf′(r′)/δωd(r′) = 0 = δf′(r′)/δQ′(r′) can be written as-∇′2ωd=2κ2[(∇y′2ωs)2κ2ωd-sd-gd-ωdQ′2-ϵv{Qx′(∇x′ωs)(Qy′(∇y′ωs)-Qx′(∇x′ωs))(∇y′2ωs)2κ2cos(ϕ)×[(-(∇y′2ωs)2κ2+(∇x′2ωs)2κ2+gsy-gsx)×(ωdωs)1/2+(Qy′2-Qx′2)(ωsωd)1/2(-(∇y′2ωs)2κ2+(∇x′2ωs)2κ2+gsy-gsx)]+2(sin(ϕ)2κ)(ωdωs)1/2×(Qy′(∇y′ωs)-Qx′(∇x′ωs))(∇y′2ωs)2κ2}(∇y′2ωs)2κ2],-∇′2ωs=2κ2[(∇y′2ωs)2κ2-ωs-ss-gs-ωsQ′2-ϵv{(Qy′(∇y′ωd)-Qx′(∇x′ωd))(∇y′2ωs)2κ2cos(ϕ)×[(-(∇y′2ωd)2κ2+(∇x′2ωd)2κ2+gdy-gdx)×(ωsωd)1/2+(Qy′2-Qx′2)(ωsωd)1/2(-(∇y′2ωd)2κ2+(∇x′2ωd)2κ2+gdy-gdx)]+2(sin(ϕ)2κ)(ωsωd)1/2×(Qy′(∇y′ωd)-Qx′(∇x′ωd))(∇y′2ωs)2κ2}(∇y′2ωs)2κ2],-∇′2Q′=-(ωs+ωd)Q′-ϵv[2cos(ϕ)(ωsωd)1/2(ŷQy′-x̂Qx′)+(sin(ϕ)2κ)×(ŷ(∇y′ωs)-x̂(∇x′ωs))(ωdωs)1/2-(sin(ϕ)2κ)(ŷ(∇y′ωd)-x̂(∇x′ωd))×(ωsωd)1/2],
where gi=(∇′ωi)2/4κ2ωi and gij=(∇j′ωi)2/4κ2ωi, with i=s,d; j=x,y; andss=t2(1-t2)(1-11+ωs(1-t4)-βsd1-ωd(1-t4)1+ωs(1-t4))ωs,sd=t2(1-t2)(11-ωd(1-t4)-1+βsd1+ωs(1-t4)1-ωd(1-t4))ωd.
The order parameters and the magnetic field are now expressed in terms of Fourier series [22], and the positions of the vortices are defined in terms of the reciprocal lattice vectors as K=Kmn=(2π/x1y2)(my2,nx1+mx2), where m and n are integers and x1, x2, and y2 are lattice parameters. The order parameter components and the magnetic field are determined by solving a set of iterative equations using a numerical iteration technique [17, 22–25]. The iteration process is continued till the solution remains constant upto 15 digits. The high precision solutions of the GL equations are thus obtained and these solutions can be used to study the various magnetic and electrostatic properties of the high-Tc superconductors involving mixed symmetry state of the order parameter components at different temperatures.
3.1. Single Vortex and Vortex Lattice Structure
The first step to study the magnetic properties of the high-Tc superconducting cuprates is to study the structure of the flux line lattice. Small angle neutron scattering (SANS) [31] and Scanning tunneling microscopy (STM) [32] experiments on high-Tc superconductors have shown that unlike the case of the conventional type-II superconductors, which are characterized by a triangular vortex lattice, the high-Tc superconductors exhibit an oblique vortex lattice configuration. The experimental observations have been substantiated by the previous theoretical works where an oblique vortex lattice structure have been observed [17]. Another important feature of the high-Tc superconductors involving mixed symmetry state of the order parameters is the fourfold symmetric structure of the s-wave order parameter component [15–17]. Figure 2 shows the vortex lattice structure of the s-wave and d-wave order parameter components corresponding to different temperatures at a particular magnetic field induction and mixed gradient coupling parameter ϵv mentioned in the figure caption. It can be seen from the figure that at different temperatures the structure of the vortex lattice is essentially oblique and the s-wave order parameter component possesses a fourfold symmetric structure. Similar oblique flux line lattice structure is observed for the magnetic field distribution also as can be seen from Figure 3.
Variation of vortex lattice structure corresponding to the s- and d-wave order parameter components ωs(x,y) ((a)–(c)) and ωd(x,y) ((d)–(f)) for different temperatures (t=0.9,0.7,0.5), respectively, calculated by the extended two-order parameter GL theory. Other parameters used for the calculation are GL parameter κ=72, magnetic field induction b=0.8, and mixed gradient coupling parameter ϵv=0.1, βsd=0.
Variation of magnetic field distribution B(x,y) for different temperatures (t=0.9,0.7,0.5) ((a)–(c)) respectively, calculated by the extended two-order parameter GL theory. Other parameters used for the calculation are same as in Figure 2.
3.2. Reversible Magnetization
An important and experimentally determinable quantity for any superconducting material is the reversible magnetization of the system. The reversible magnetization is defined as M=B¯-Ba, where Ba is the equilibrium applied magnetic field and is expressed as Ba=4π(∂f/∂B¯). Thus, the determination of the equilibrium applied magnetic field of the system involves the computation of the numerical derivative of the free energy density functional which is a complex function of two-order parameter components and magnetic field. An alternative approach is to determine the equilibrium applied magnetic field by making use of the virial theorem applicable to the two-order parameter model [22]. In the framework of the two-order parameter extended GL theory, the virial theorem has been formulated and the resulting equilibrium applied magnetic field can be expressed asBa=〈2B2+ωd-ωs-sd-ss〉2B¯.
Using this expression for the equilibrium-applied magnetic field the corresponding reversible magnetization of the system can be calculated as per the relation M=B¯-Ba. The reversible magnetization of the system calculated by this expression at different temperatures is plotted in Figure 4. It can be seen from the figure that for higher values of t, that is, at temperatures close to Td, the reversible magnetization shows the behavior well known for the GL theory. However, as one move to lower temperatures an anomalous behavior of the reversible magnetization is observed, characterized by an s-shape of the curve whose curvature increases with the decrease in temperature. In case of conventional type-II superconductors also the curvature has been found to increase with the decrease in the temperature and below a certain temperature termed as Ta the reversible magnetization shows an anomalous behavior and the system undergoes a first-order transition [13, 14]. At temperatures below Ta, a finite magnetization is observed for an applied magnetic field above Bc2. For the conventional type-II superconductors with κ≈1/2, a first-order transition has been predicted near the Hc1 and Hc2, depending upon the ratio of the mean free path l to the coherence length ξ0. For the conventional low temperature type-II superconductors, the anomalous behavior of the reversible magnetization of the system has been attributed to the presence of impurities in the superconducting system [33, 34].
Applied magnetic field Ba (measured in units of upper critical field Bc2) dependence of reversible magnetization calculated by the extended two-order parameter GL theory for different temperatures. Other parameters used in the calculation are GL parameter κ=72 and mixed gradient coupling parameters ϵv=0.1 and βsd=0.
In case of the high-temperature superconductors involving mixed symmetry state of the order parameters, such a first-order transition has not been observed for the temperatures plotted in Figure 4. The absence of such a transition can be attributed to the large value of the GL parameter κ in case of the high-Tc superconductors. Before coming to this point, it will be useful to determine the temperature Ta in case of the high-Tc superconductors. Near the upper critical field Bc2 the density of superconducting condensate is small and the coefficients of the GL free energy functional can be defined asα=γ2n(T2-Td2),β=γT22n2.
We now define an asymptotic GL parameter κas as [35]κas=md2β2μ0ℏ2e2.
Thus, the asymptotic GL parameter κas is related to temperature as per the relationκas=κ0t2.
The first-order transition should be observed for κas=1/2 and with κ0=72 corresponding to high-Tc superconductor YBa2Cu3O7 − δ, the transition temperature (Ta) amounts toTa=2κ0Td=0.01964.
Thus, in case of high-Tc superconductors (particularly YBa2Cu3O7 − δ) such a transition is likely to be observed at a very low temperature and the range of validity of the extended GL theory for high-Tc superconductors is larger as compared to that for the conventional type-II superconductors. This justifies the observation that in Figure 4 a finite reversible magnetization is not observed at applied magnetic fields higher than Bc2 even at a low temperature of t=0.4. One may possibly define as in the case of the applied magnetic field (Hc1<H<Hc2), a range of temperature as Ta<T<Tc, over which the extended GL theory is valid for a type-II superconducting material. Below this temperature Ta, one should be careful regarding the validity of the extended GL theory and the accuracy of the results obtained.
4. Electrostatic Potential and Charge Distribution in Flux Line Lattice
We next concentrate on the electrostatic potential and the associated charge distribution in the vortices of the high-Tc superconductors involving mixed symmetry state of the order parameter components.
4.1. Electrostatic Potential
The electrostatic potential of the system is expressed as per equation (13). For the sake of simplicity, in the present calculation of the electrostatic potential the Thomas-Fermi screening is neglected, that is, λTF2∇2eϕ=0. Such an approximation is justified since the Thomas-Fermi screening length is small as has been shown below. In terms of dimensionless units, we define the electrostatic potential asΦ=en(1/4)γTd2(1-t2)(1-t4)ϕ.
For the high-Tc superconductors involving mixed symmetry state of the order parameter components, the resulting electrostatic potential isΦ=(ωs-ωd)+(ss+sd)+C1(ωd-ωs)+C2(1+ωs(1-t4)+1-ωd(1-t4)+βsd1+ωs(1-t4)1-ωd(1-t4)),
where the temperature-dependent coefficients C1 and C2 are given asC1=1(1-t2)∂lnϵcon∂lnn,C2=2t2(1-t2)(1-t4)∂lnγ∂lnn.
The coefficients C1 and C2 depend upon the material parameter of the system under consideration, in this case YBCO. Details regarding the determination of the coefficients C1 and C2 for YBCO are discussed in the appendix at the end of the paper.
The electrostatic potential expressed by (27) consists of three components as Φ1=C1(ωd-ωs), which arises due to the dependence of the condensation energy on the superconducting electron density, Φ2=C2(1+ωs(1-t4)+1-ωd(1-t4)+βsd1+ωs(1-t4)1-ωd(1-t4))arising from the reduced normal-state thermoelectric potential and ΦB=(ωs-ωd)+(ss+sd) corresponding to the Bernoulli potential. The individual potential components are shown in Figure 5 for the temperature t and magnetic field induction b mentioned in the figure caption. A characteristic feature that can be observed from the figure is the flatness of the curves corresponding to the different components of potential, this is, unlike the case observed for conventional type-II superconductor niobium [13, 14]. It is worth mentioning that Kumagai et al. [12] in their NQR measurement of vortex charge accumulation for YBa2Cu3O7 and YBa2Cu4O8 have observed a flat response of the vortex charge distribution. One can thus predict that the corresponding electrostatic potential should also show similar flatness in their profiles.
The components of the electrostatic potential Φ plotted along the x-direction. The parameter values used for the figure are t=0.5, magnetic induction parameter b=0.5, and GL parameters κ=72 and βsd=0.
The Bernoulli potential ϕB reaches zero at the center of the vortex. The potential is repulsive inside the vortex core, while outside the core the Bernoulli potential is attractive. Similar to the Bernoulli potential ϕB, the ϕ1 component of electrostatic potential minimizes at the center of the vortex. This is an expected observation, since ϕ1 arises from the condensation energy and is thus proportional to the density of superconducting electrons. The density of superconducting electron density vanishes at the center of the vortex and so is the corresponding component of the electrostatic potential. The third component of the electrostatic potential ϕ2 is the only component which gives a nonzero contribution at the center of the vortex.
4.2. Electrostatic Charge Distribution
We next calculate the distribution of electrostatic charge in the flux line lattice. As per the Poisson's equation, the charge accumulation in the vortex core can be calculated from the electrostatic potential of the system as per the relation ρ=-ϵ∇2ϕ. We define charge in terms of dimensionless units asρ′=ρen.
The corresponding Poisson's equation in terms of the dimensionless unit is given asρ′=-C3λTF2λLon2∇′2Φ,
where the coefficient C3 is dependent upon the material parameters of the superconducting system under consideration, in this case YBCO, and is expressed asC3=2Dϵconn2(1-t2)(1-t4)
with D being the density of states. As in case of the electrostatic potential of the vortex lattice, the net electrostatic charge in the vortex lattice also consists of three parts as ρ1′, ρ2′, and ρB′ corresponding to the electrostatic potential components ϕ1, ϕ2, and ϕB, respectively.
The different components of charge are plotted in Figure 6, and the particular shape of the vortex charge profiles are strongly dependent on the material parameters of YBCO through the coefficients C1, C2, and C3. At the center of the vortex, the components of charge ρB and ρ2 give a positive contribution while the magnitude of the component ρ1 is negative at the center of the vortex. The net charge at the center of the vortex has a finite negative value due to the large negative contribution arising from the component ρ1. The total charge density (ρ) shows a flat response outside the vortex. The behavior can be considered to be in agreement with the experimental observation reported by Kumagai et al. for YBCO [12].
Different components of the charge distribution profile calculated by two-order parameter extended GL theory. The parameter values are the same as in Figure 5.
For the calculation of charge distribution in the vortex lattice, the effect of Thomas-Fermi screening has been taken into account. It must however be noted that the magnitude of screening is very small. In case of YBCO, this screening amounts toλTF2λLon2=1.4646×10-7
with λTF2=ϵ/2De2 and λLon=1.4×10-6m. Consequently, for YBCO the coefficientC3λTF2λLon2=1.54×10-9(1-t2)(1-t4).
In Figure 7, the magnetic field dependence of peak amplitude of charge calculated at different temperatures is presented. A higher applied magnetic field corresponds to a lower amplitude of charge. At higher magnetic field inductions, the peak amplitude of charge almost saturates. A possible reason for the observation can be the greater overlap between the vortices at higher magnetic field. At high magnetic field, the charge accumulation in the vortex decreases along the line joining the neighboring vortices. A lower temperature corresponds to a greater accumulation of charges in the vortex core.
Magnetic field dependence of peak amplitude of charge ρ(x,0) calculated along the x-direction at different temperatures. Other relevant parameters are κ=72, ϵv=0.1, and βsd=0.
5. Conclusions
The study carried out in the present work consists of two parts, the first part is the extension of the GL theory to the low temperature regime in case of the high-Tc superconducting cuprates involving mixed symmetry state of the order parameter components. The second part of the work involves the study of the electrostatic potential and the associated charge distribution in the vortices. In order to study the electrostatic potential and charge accumulation in the vortices, a set of equations are derived which consist of the Poisson's equation for electrostatic charge, Bernoulli potential for the electrostatic potential, Ginzburg-Landau equation for s- and d-wave order parameter components, and Ampere's law for magnetic vector potential. The resulting equations are solved by using a numerical iteration technique for arbitrary magnetic field induction and wide range of temperature. At any temperature, the flux line lattice shows an oblique structure characteristic to the high-Tc cuprates.
The equilibrium applied magnetic field and the resulting reversible magnetization is calculated by using the virial theorem developed for the two-order parameter system. The reversible magnetization could thus be obtained without taking the numerical derivative of the free energy density functional. In the applied magnetic field versus reversible magnetization plot, an interesting S-shaped feature has been observed. The curvature of the S-shaped curve has been found to get enhanced at the lower values of temperature; however, a first-order transition, as has been found in case of conventional type-II superconductors, is absent in case of the high-Tc superconducting cuprates for the temperature region studied. It can be attributed to the value of the temperature Ta, which determines the lower limit of applicability of the extended GL theory. In case of high-Tc superconducting cuprates, the value of Ta is much lower than that in case of the conventional type-II superconductors and thus any anomalous feature of the reversible magnetization curve can be expected to be observed only below this temperature. The observation also signifies that as in case of the applied magnetic field (Hc1<H<Hc2), for temperature also it is possible to define a range (Ta<T<Tc) over which the extended GL theory is applicable. In case of high-Tc superconducting cuprates, the extended GL theory is valid over a wider temperature regime as compared to the conventional low-temperature type-II superconductors.
The work further deals with the determination of the electrostatic potential and the associated charge distribution in the vortices. The net electrostatic potential and charge distribution in the vortex lattice consists of three different components and their magnitudes are strongly dependent on the material parameters. In case of the high-Tc cuprate YBCO, the spatial distribution of the electrostatic potential and charge shows a flat behavior unlike the case of the conventional low-temperature type-II superconductors. Here, it is worth mentioning that Kumagai et al. [12] through their experimental study have indeed suggested a flatness in the vortex charge distribution profile in high-Tc superconducting cuprates.
A correspondence between the theoretical results and experimental observations can be achieved by utilizing the present theoretical model to explain the result obtained from the NMR and NQR studies carried out on the high-Tc superconducting cuprates. The present model can further be generalized to take into account the various important features associated with the high-Tc cuprates such as, the presence of anisotropy or orthorhombic distortion. Apart from the high-Tc cuprates, the method can be used to analyze the electrostatic potential and charge distribution in other materials involving complex order parameters. These issues will be addressed in future.
AppendixA. Calculation of Material Parameters for YBCO
In this section, we determine the various material parameters of high-Tc cuprate YBCO that has been used to calculate the coefficients C1 and C2.
A.1. Calculation of ∂γ/∂n
The linear coefficient of specific heat γ is related to the density of states D as per the relationγ=23π2kB2D.
Here, the linear coefficient of specific heat associated with the superconducting CuO2-plane of YBCO has been taken into account. The effect of the chains has not been considered. For the high-Tc cuprate YBa2Cu3O7, the linear coefficient of specific heat is chosen to be 302JK-2m-3 [36]. Thus, the corresponding density of states can be determined by the relationD=3γ2π2kB2.
The density of states D, which includes the mass renormalization due to the electron-phonon coupling, is related to the bare density of states D0 asD=D0(1+λ),
where λ is the electron-phonon coupling constant. As per the Kresin-Wolf two-gap theory [37–40], in case of the high-Tc superconductors involving mixed symmetry state of order parameters, λ is a mixture of the coupling constants corresponding to superconductivity in the CuO2-planes, in the CuO-chains and the interaction between the plane and the chain. The plane exhibits strong-coupling superconductivity characterized by the coupling constant λ=3. The contribution to λ arising from the other two sources mentioned above ranges from 0.5 to 0.9. In the present study, a two-dimensional approach to the problem is implemented wherein the superconducting property of the high-Tc cuprates is considered to be originating from the CuO2-plane. Both the dominant d-wave and the subdominant s-wave order parameter components are considered to be residing in the CuO2-plane, as has been interpreted from the recent experimental study carried out on YBCO [18, 19]. Consequent to the two-dimensional approach to the problem, in the present study the electron-phonon coupling constant is considered to be λ=3.
From the bare density of states, one can determine the corresponding BCS interaction (V) as per the expressionλ=D0V.
Using (A.1) the density derivative of the linear coefficient of specific heat can be expressed as∂γ∂n=13π2kB2∂lnD∂EF,
where we have used ∂EF/∂n=1/2D and n=5×1027m-3 is the density of holes [36]. Using (A.2–A.4), the density derivative of the linear coefficient of specific heat can be expressed as∂γ∂n=13π2kB21+2λ1+λ∂lnD0∂EF,
where the BCS interaction V has been taken to be constant, that is, ∂V/∂n=0 or ∂V/∂EF=0.
Using (A.6), the coefficient ∂γ/∂n can be determined. The energy derivative of the bare density of state ∂D0/∂EF is obtained from the experimental data [41].
A.2. Calculation of ∂ϵcon/∂n
For determining ∂ϵcon/∂n, we begin by expressing the critical temperature (in the present case Td) in terms of McMillan formula [42]Td=ΘD1.45exp[-1.04(1+λ)λ-μ*(1+0.62λ)],
where ΘD=440K [43] is the Debye temperature and μ*=0.2 is the Coulomb pseudopotential. The corresponding condensation energy can thus be expressed asϵcon=π212.6kB2(1+λ)D0ΘD2exp[-21.04(1+λ)λ-μ*(1+0.62λ)],
where we have used the relation ϵcon=(1/4)γTd2. The density derivative of the condensation energy can thus be given as∂ϵcon∂n=ϵcon∂λ∂n[2.08(1+0.38μ*)[λ-μ*(1+0.62λ)]2+11+λ],
where we have considered the product D0ΘD2 and μ* to be constant with respect to the variation in the density of hole n.
Using the relations between λ and D0 given by (A.2–A.4) one can write∂λ∂n=V2(1+λ)∂lnD0∂EF,
where the energy derivative of the bare density of state is again determined from experimental data for YBCO [41]. The resulting expression for the density derivative of the condensation energy is thus given as∂ϵcon∂n=ϵconV(1+λ)2∂lnD0∂EF(1.04(1+0.38μ*)(1+λ)[λ-μ*(1+0.62λ)]2+12).
Using (A.11) one can calculate the coefficient ∂ϵcon/∂n. In Table 1, we have tabulated the important material parameters for YBCO that has been used for the present study.
Material parameters for YBCO.
Critical temperature
Tc
90 K
Debye temperature
ΘD
440 K
Coupling parameter
λ
3
Coulomb pseudo potential
μ*
0.2
Coefficient of specific heat
γ
302JK-2m-3
Hole density
n
5×1027m-3
Logarithmic derivative
∂D0/∂EF
7.5×1019J-1
Density of states
D
2.41×1047J-1m-3
Bare density
D0
0.602×1047J-1m-3
BCS interaction
V
4.98×10-47Jm3
Condensation energy
ϵcon
5.45×105Jm-3
Coefficient (A6)
∂γ/∂n
8.23×10-26J
Coefficient (A11)
∂ϵcon/∂n
16.02×10-23J
Coefficient of C1
∂lnϵcon/∂lnn
1.47
Coefficient of C2
∂lnγ/∂lnn
1.4
Acknowledgments
The author gratefully acknowledges E. H. Brandt and Pavel Lipavský for fruitful discussions. The work is financially supported by S. N. Bose National Centre for Basic Sciences, India.
KhomskiiD. I.FreimuthA.Charged vortices in high temperature superconductors1995757138413862-s2.0-424378396010.1103/PhysRevLett.75.1384BlatterG.Feigel'manM.GeshkenbeinV.LarkinA.van OtterloA.Electrostatics of vortices in type-II superconductors1996773566569HayashiH.IchiokaM.MachidaK.Relation between Vortex core charge and Vortex Bound States1998671033683371BoppF.über die beziehungen der Londonschen gleichungen zur beschleunigungstheorie der supraleitung19371079-106236322-s2.0-000047040410.1007/BF01339298BokJ.KleinJ.‘Electric Fields’ in superconductors196820136606622-s2.0-334296575310.1103/PhysRevLett.20.660BrownJ. B.MorrisT. D.AllenJ. F.FinlaysonD. M.McCallD. M.2Proceedings of 11th International Conference on Low Temperature PhysicsAugust 1968ScotlandSt. Andrews768HongK. M.Bernoulli potential in superconductors19751251766177110.1103/PhysRevB.12.1766LondonF.1950ISec. 8New York, NY, USAWileyAdkinsC. J.WaldramJ. R.Bernoulli potentials in superconductors196821276782-s2.0-424378396210.1103/PhysRevLett.21.76RickayzenG.Bernoulli potentials in superconductors196927133413372-s2.0-0014539978NagaokaT.MatsudaY.ObaraH.SawaA.TerashimaT.ChongI.TakanoM.SuzukiM.Hall anomaly in the superconducting state of high-Tc cuprates: universality in doping dependence19988016359435972-s2.0-0032049931KumagaiK.NozakiK.MatsudaY.Charged vortices in high-temperature superconductors probed by NMR2001631452-s2.0-0034904747144502LipavskýP.KoláčekJ.MorawetzK.BrandtE. H.Electrostatic potential in a superconductor2002651418144511LipavskýP.KoláčekJ.MorawetzK.BrandtE. H.YangT.-J.2008733Berlin, GermanySpringerLecture Notes in PhysicsFranzM.KallinC.SoininenP. I.BerlinskyA. J.FetterA. L.Vortex state in a d-wave superconductor1996539579558142-s2.0-0000446678HeebR.van OtterloA.SigristM.BlatterG.Vortices in d-wave superconductors19965413938593982-s2.0-0001188217KarmakarM.madhu@physics.unipune.ernet.inDeyB.bdey@physics.unipune.ernet.inProperties of high temperature superconductors in states of mixed symmetry2008202525521810.1088/0953-8984/20/25/255218KhasanovR.SträssleS.Di CastroD.MasuiT.MiyasakaS.TajimaS.Bussmann-HolderA.KellerH.Multiple gap symmetries for the order parameter of cuprate superconductors from penetration depth measurements20079923423760110.1103/PhysRevLett.99.237601KhasanovR.ShengelayaA.MaisuradzeA.La MattinaF.Bussmann-HolderA.KellerH.MüllerK. A.Experimental evidence for rwo gaps in the high-temperature La1.83Sr0.17CuO4 superconductor200798542-s2.0-3384682202010.1103/PhysRevLett.98.057007057007SmildeH. J. H.GolubovA. A.AriandoRijndersG.DekkersJ. M.HarkemaS.BlankD. H. A.RogallaH.HilgenkampoH.Admixtures to d-wave gap symmetry in untwinned YBa2Cu3O7 superconducting films measured by angle-resolved electron tunneling20059525425700110.1103/PhysRevLett.95.257001AffleckI.FranzM.SharifzadehM. H.Generalized London free energy for high-Tc vortex lattices1997552R704R7072-s2.0-0000041834KarmakarM.DeyB.Reversible magnetization of the layered high-Tc superconductors20067417417250810.1103/PhysRevB.74.172508KarmakarM.madhu@physics.unipune.ernet.inPairing state symmetries of high-Tc superconductors: a comparative study using two Ginzburg-Landau models2009211717570210.1088/0953-8984/21/17/175702KarmakarM.madhu@physics.unipune.ernet.inDeyB.bdey@physics.unipune.ernet.inThe effect of in-plane mass anisotropy on the properties of high temperature superconductors in states of mixed symmetry2009214040570210.1088/0953-8984/21/40/405702KarmakarM.madhu@physics.unipune.ernet.inAnomalous magnetic field dependence of y Ba2Cu3O7−δ: effect of orthorhombic distortion: effect of orthorhombic distortion20102225 25570110.1088/0953-8984/22/25/255701BardeenJ.Theory of boundary effects of superconductors19549435545632-s2.0-3614901070710.1103/PhysRev.94.554GorterC. J.CasimirH. B. G.193435963RenY.XuJ.-H.TingC. S.Ginzburg-Landau equations and vortex structure of a dx2-y2 superconductor199574183680368310.1103/PhysRevLett.74.3680XuJ. -H.RenY.TingC. S.Ginzburg-Landau equations for a d-wave superconductor with applications to vortex structure and surface problems199552107663767410.1103/PhysRevB.52.7663FederD. L.KallinC.Microscopic derivation of the Ginzburg-Landau equations for a d-wave superconductor19975515595742-s2.0-0000483724KeimerB.ShihW. Y.ErwinR. W.LynnJ. W.DoganF.AksayI. A.Vortex lattice symmetry and electronic structure in YBa2Cu3O719947325345934622-s2.0-064231045910.1103/PhysRevLett.73.3459Maggio-AprileI.RennerC.ErbA.WalkerE.FischerO.Direct vortex lattice imaging and tunneling spectroscopy of flux lines on YBa2Cu3O7-δ19957514275427572-s2.0-034317847210.1103/PhysRevLett.75.2754JacobsA. E.First-order transitions at hc1 and hc2 in type-II superconductors197126116296312-s2.0-3374455785510.1103/PhysRevLett.26.629JacobsA. E.First-order transitions at hc1 and hc2 in type-II superconductors197149302230292-s2.0-424418235610.1103/PhysRevB.4.3022WaldramJ. R.1996Bristol, UKArrowsmithPlakidaN. M.1995Berlin, GermanySpringerKresinV. Z.WolfS. A.Multigap structure in the cuprates19901695-6476484KresinV. Z.WolfS. A.Major normal and superconducting parameters of high-Tc oxides19904174278428510.1103/PhysRevB.41.4278KresinV. Z.WolfS. A.Induced superconducting state and two-gap structure: application to cuprate superconductors and conventional multilayers199246106458647110.1103/PhysRevB.46.6458KresinV. Z.On the critical temperature for any strength of the electron-phonon coupling19871228434438SchwingenschlöglU.Udo.Schwingenschloegl@physik.uni-augsburg.deSchusterC.Intrinsic doping at YBCO-metal interfaces: quantitative results20077733700710.1209/0295-5075/77/37007KochC. C.ScarbroughJ. O.KroegerD. M.Effects of interstitial oxygen on the superconductivity of niobium1974938888972-s2.0-000125977810.1103/PhysRevB.9.888InderheesS.SalamonM.FriedmannT.GinsbergD.Measurement of the specific-heat anomaly at the superconducting transition of YBa2Cu3O7-δ1987364240124032-s2.0-000155855710.1103/PhysRevB.36.2401