PHYSICAL REVIEW B 108, L020502 (2023)

Letter

Particle-hole spectral asymmetry at the edge of multiorbital noncentrosymmetric superconductors

Yuri Fukaya ,1 Keiji Yada,2 Yukio Tanaka,2 Paola Gentile,1 and Mario Cuoco1
1
CNR-SPIN, c/o Universitá di Salerno, I-84084 Fisciano (Salerno), Italy
2
Department of Applied Physics, Nagoya University, Nagoya 464-8603, Japan

(Received 3 October 2022; accepted 22 June 2023; published 6 July 2023)

Bogoliubov quasiparticles are a coherent electron-hole quantum superposition which typically, for time-
reversal symmetric superconductors, exhibit a spectral distribution with particle-hole symmetry. Here, we
demonstrate that in two-dimensional noncentrosymmetric superconductors with multiorbital spin-triplet pairing
the energy profile of the density of states at the edge can violate this paradigm. We show that the structure of
Andreev reflections generally leads to pairing states made of configurations that are orbitally split due to the low
degree of crystalline symmetry at the edge. The resulting pairing state has a mixed parity character in the orbital
sector that, in the presence of reduced crystal symmetry at the edge, sets out a particle-hole asymmetric profile
for the spectral function. These findings indicate a path to design asymmetric spectral functions at the edge of
superconductors with orbital degrees of freedom and time-reversal symmetry. The emerging signatures can be
exploited for the detection of spin-triplet pairing equipped with internal degrees of freedom.

DOI: 10.1103/PhysRevB.108.L020502

Introduction. Most of the quantum coherent properties of commodate Andreev bound states that lead to a particle-hole
superconductors get manifested when the superconducting asymmetric profile of the density of states? In this Letter, we
state is varying in space, as for instance at the edge, nearby provide an answer to this problem. We demonstrate that in
the surface, or in proximity to another system. In spatially two-dimensional noncentrosymmetric superconductors with
inhomogeneous regions, indeed, multiple Andreev reflections multiorbital s-wave spin-triplet pairing the density of states
[1] lead to the formation of Andreev bound states with char- at the edge is generally marked by an asymmetric line shape.
acteristic excitation energies below the superconducting gap We show that the lack of horizontal mirror symmetry induces
[1–6]. The observation of these states, e.g., through scanning a distinct type of orbital-dependent Andreev reflections, via
tunneling microscopy (STM) or angle-resolved photoemis- the so-called orbital Rashba interaction [30–34], that is able
sion spectroscopy (ARPES), is crucial to assess the nature to set out a pairing at the edge with orbital mixed parity in the
of the superconducting phase and of the underlying pairing spin-triplet channel. This pairing state indeed includes config-
symmetry [3,6]. This is particularly challenging and relevant urations with orbital singlet (L = 0) and orbital quintet (L =
in superconductors having topologically protected boundary 2) angular momentum (or equivalently mixed singlet-triplet
modes that are robust with respect to local perturbations configurations in the pseudospin manifold spanned by two
[6–9]. An emblematic case in this context is provided by the orbitals). We demonstrate that this type of pairing in the pres-
Majorana zero energy modes, as their successful generation ence of the orbitally driven interactions that reduce the crystal
and control represent a fundamental milestone for achieving symmetry at the edge of the superconductor can generally
topologically protected quantum computation [10]. yield asymmetric spectral distribution of the density of states.
Recently, STM investigations of the just discovered Model. To illustrate this physical scenario, we consider a mul-
heavy-fermion superconductor UTe2 [11–13] revealed an tiorbital 2D electronic system with spin-triplet s-wave pairing
asymmetric line shape of the density of states at surface and broken inversion symmetry. The electronic structure is
step edges with features that break the particle-hole symme- marked by three bands arising from atomic states spanning an
try expected for Bogoliubov quasiparticles [14]. Moreover, L = 1 angular momentum subspace, such as da orbitals with
the energy asymmetric peaks that are observed inside the a = (yz, zx, xy). Here, we refer to d orbitals localized at the
superconducting gap exhibit a behavior that depends on site of a square lattice assuming a C4v point group symmetry.
the direction of the normal to the side surface of the step edge The breaking of planar mirror symmetry (Mz ) sets out a polar
[14]. It is known that the particle-hole asymmetry in the super- axis, z, leading to an orbital Rashba interaction (αOR ) that
conducting spectra can descend from the normal state [15–17] mixes orbital configurations with different mirror parity, i.e.,
or due to pair breaking scattering [18], as it occurs in cuprates xy with xz or yz. As for the spin Rashba interaction, the orbital
due to their large energy gap. Similar asymmetric features Rashba couples the atomic angular momentum L with the
are also expected in the presence of sources of time-reversal crystal wave vector k in the usual form, i.e., αOR [L̂x sin(ky ) −
symmetry breaking [19–24] as well as due to mechanisms L̂y sin(kx )] [30–34]. In the manifold of the da orbitals, the
driven by electron-boson interactions [25–27] and dissipative components of the angular momentum can be expressed in a
coupling [28,29]. matrix form as [L̂k ]lm = iklm with klm the Levi-Civita tensor,
In this context, a fundamental and general question arises: while hereafter ŝ j=x,y,z refers to the Pauli matrix in spin space,
What type of time-reversal symmetric superconductor can ac- with ŝ0 being the identity matrix. Assuming that the basis

2469-9950/2023/108(2)/L020502(8) L020502-1 ©2023 American Physical Society

FUKAYA, YADA, TANAKA, GENTILE, AND CUOCO PHYSICAL REVIEW B 108, L020502 (2023)

FIG. 1. (a) Sketch of the interorbital spin-triplet/orbital-singlet/s-wave pairing (B1 symmetry for the C4v crystal point group). The pairing
is described by two d-vectors: d y for the (xy, yz) orbital channels and d x for the (xy, zx) sector. (b), (c) The lowest-order Andreev reflection
(AR) that originates from the interorbital spin-triplet pairing in (a) at the normal metal (N)/superconductor (S) interface. Due to the d-vector
structure, as allowed by the crystal symmetry, AR for [(yz, e), (xy, h)] in (b) is not equivalent to that for [(zx, e), (xy, h)] in (c). Here, e (h)
denotes the electron (hole) quasiparticles. (d)–(g) Higher-order Andreev reflection caused by OR for (d) [(yz, e), (yz, h)], (e) [(zx, e), (zx, h)],
(f) [(yz, e), (zx, h)], and (g) [(zx, e), (yz, h)] electronic states, respectively. Ĥ0 and Ĥis indicate the scattering processes involving the hopping
term and the OR coupling, respectively. Since the pair potential is associated to spin-triplet pairs, for injected electrons and reflected holes, the
spin orientation is not modified, i.e., the relevant AR processes are spin conserving.

of the local creation operator of electrons for each d orbital to local electron pairs between {dzx , dxy } ({dzy , dxy }) orbitals,
is Ĉk† = [cyz,↑k
†
, czx,↑k
†
, cxy,↑k
†
, cyz,↓k
†
, czx,↓k
†
, cxy,↓k
†
], the Hamil- respectively. We recall that the dx and dy components refer to
 † spin-triplet pairs having zero spin projections along the x and
tonian can be generally expressed as Ĥ = k Ĉk Ĥ (k)Ĉk with
Ĥ(k) given by y directions, respectively. Due to the C4v symmetry, dx(xy,zx)
 and dy(xy,yz) are equal in amplitude. A schematic illustration of
Ĥ(k) = [a (k)Pa + αOR (gk × L̂) · ẑ − μ] ⊗ ŝ0 the spin and orbital structure of the B1 Cooper pair is shown in
a Fig. 1(a). We notice that for this configuration the pair poten-
+ λSO L̂ · ŝ, tial in the bulk does not involve all the orbitals as electrons in
the xz and yz orbitals are not directly paired. Hence it allows
where gk = [sin(kx ), sin(ky ), 0], while a (k) = us to investigate the orbital reconstruction of the pairing at the
−2tax cos(kx ) − 2tay cos(ky ) − δa indicate the dispersion edge. The B1 phase exhibits nodal points along the diagonal of
relations for each orbital a. The nearest-neighbor hopping the Brillouin zone ([110] direction) with nonzero topological
amplitudes are assumed to be tyzy = txz x
= txy
x
= txyy
= t, number, due to the chiral symmetry of the BdG Hamiltonian
y 
tyz = txz = t , and Pa = (L̂ − L̂a )/2 is a projector on the
x 2 2 [35–42], and can lead to anomalous Josephson effects [37,42].
orbital a subspace. At the  point of the Brillouin zone, a The A1 pairing configuration instead yields a fully gapped
degeneracy occurs between the xz and the yz orbitals; thus superconducting phase. Additionally, the A1 s-wave d-vector
δxz = δyz = 0. In turn, the orbital xy has a nonvanishing components can be nonvanishing for any given pair of orbitals
crystal field splitting, δxy ≡ t , whose amplitude depends in the (zx, yz, xy) manifold (see Supplemental Material [43]).
on the crystalline distortions. λSO is the atomic spin-orbit The comparison of the A1 and B1 phases, hence, allows us
coupling expressing the interaction between the spin and to discern the role of topological nodal points and the orbital
angular momentum at each site. dependence of the d-vector components.
One can tune the number of Fermi lines from two to six by Orbital dependent Andreev reflection and anomalous pair
varying the chemical potential μ. Here, we focus on the con- correlations. Andreev [1] considered the reflection of elec-
figuration with six Fermi lines, by choosing a representative trons incident from a normal region into a superconducting
value for μ, i.e., μ/t = 0.35, because we aim to have all the domain of the same material as schematically described in
orbitals active at the Fermi level. Amplitude variations of μ Fig. 1(b). For energies below the gap, the electron is forbidden
that keep the number of Fermi lines do not alter the qualitative to propagate into the superconducting region and the scat-
outcome of the results. For the superconducting state, the tering from the pair potential in the superconductor converts
Bogoliubov–de Gennes (BdG) Hamiltonian is then directly an electron into a hole. When dealing with the spin-triplet
constructed by including the pair potential (k).
ˆ We con- superconductor upon examination the Andreev reflection has
sider different symmetry allowed local (s-wave) spin-triplet an inherently richer spin-orbital structure [Figs. 1(b) and 1(c)].
pairing with orbital-singlet character. The various irreducible Indeed, a spin polarized electronlike excitation with, for in-
representations can involve s-wave (k-independent) d-vector stance, xz orbital character upon reflection is converted into
components for any given pair of orbitals [35]. The focus here an outgoing xy hole excitation with the same spin orienta-
is on superconducting states with B1 and A1 symmetry in the tion. This reflection involves the dy spin-triplet pair potential
C4v group. The B1 order parameter is described [35–37] by [Fig. 1(b)]. A similar process occurs for the zx electronlike
an s-wave d vector with dx (dy ) components corresponding excitation that, via the dx pair potential, gets converted into

L020502-2
PARTICLE-HOLE SPECTRAL ASYMMETRY AT THE EDGE … PHYSICAL REVIEW B 108, L020502 (2023)

an outgoing xy hole excitation [Fig. 1(c)]. We thus observe

that basic Andreev reflections involve conversion among or-
bital states with opposite Mz mirror parity. However, we note
that the orbital Rashba coupling involves charge transfers
along the x and y directions [Fig. 1(d)] with orbital hybridiza-
tion that does not conserve the Mz mirror parity. This aspect
is thus relevant to allow for high-order Andreev reflections
that can convert electron into hole states with the same hori-
zontal mirror parity. The resulting processes are schematically
described in Figs. 1(e) and 1(f). Indeed, multiple reflections
by the superconducting pair potential combined with inter-
mediate orbital mixing, through the orbital-Rashba coupling
(His ), leads to both yz-electron to yz-hole and yz-electron
to zx-hole conversion. We observe that the interorbital yz-zx
reflection requires both d x and d y pair potentials; thus an
effective rotation in the spin space of the spin-triplet d vector
is needed. On the other hand, the equal orbital (i.e., zx or yz)
electron-hole conversion is achieved without rotating the d
vector. These processes indicate that the induced intra- and
interorbital pairing amplitudes are generally not equivalent,
thus leading to an orbital mixed parity configuration.
In order to verify and quantify the orbital structure of the
pair correlations at the edge we evaluate the corresponding
anomalous Green’s function. The analysis is performed by
means of the recursive Green’s function method [44–46] as- FIG. 2. Anomalous part of the retarded Green’s function
suming a semi-infinite geometry for the superconductor. The F̂ (E , ky ) for the B1 representation at a representative momentum
anomalous Green’s function, F̂ (E , ky ), is then evaluated for ky = 0.05π at the (100) termination for the equal-spin intraorbital
equal spin orientation, i.e., ↑↑ and ↓↓, and within the (zx, yz) channels (a) (yz ↑, yz ↑) and (b) (zx ↑, zx ↑) and equal-spin interor-
orbital manifold, at the edge of the superconductor, assuming bital components (c) (yz ↑, zx ↑) and (d) (zx ↑, yz ↑). Red-solid
that the termination is perpendicular to the (100) direction. (blue-dotted) line denotes the real (imaginary) part of the anomalous
The results demonstrate that the edge pair correlations in- part. At finite momentum the real part of the anomalous Green’s
duced in the (xz, yz) manifold for spin-triplet pairs have a rich function has both even and odd frequency components. The orbital
orbital structure (Fig. 2). At zero momentum the real part of Rashba interaction is αOR /t = 0.20, the chemical potential μ/t =
the anomalous Green’s function is purely odd in frequency 0.35, and the gap amplitude |0 | = 0.0010t, respectively.
[47–54], while at finite momentum it exhibits both even and
odd frequency components. This is expected on the basis of superconductor by considering a 2D geometry with different
the Fermi-Dirac statistics and the multiorbital character of terminations [i.e., (100) and (110)] and for the B1 and A1
the pairing amplitude. Moreover, we observe that, due to the pairing symmetry in the bulk. The density of states (DOS)
matrix structure of L̂, the eigenstates of its components L̂k , D(E ) is calculated by the retarded Green’s function G̃(E , ky ),
with nonvanishing quantum numbers (+1, −1), are generally 
expressed as combinations of two out of three states with zero 1
D(E ) = − Im dky Tr  [G̃(E , ky )]. (1)
orbital polarization. Hence one has, for instance, that for Lz π
the (+1, −1) configurations are given by |±z = √12 (∓i|xz +
In the superconducting state, the trace is performed by sin-
|yz). Then, in the (xz, yz) orbital manifold, since the intra- gling out the electron contribution. We adopt the recursive
and interorbital anomalous Green’s functions are nonvanish- Green’s function method [45] to obtain the retarded Green’s
ing, one can have pairs with total orbital angular momentum function G̃(E , ky ) at a surface layer (see Supplemental Mate-
Lz = 2, i.e., with +z +z and −z −z configurations. Taking into rial [43]).
account that for the spin channel the pairs have a spin-triplet As mentioned before, the bulk DOS for the B1 [Fig. 3(a)]
structure, and by combining the spin and orbital angular com- and A1 [Fig. 3(b)] indicates that the B1 is nodal, while the
ponents, Jz = Lz + Sz , one ends up with an effective pairing A1 has a small gap. The evaluation of the DOS at the edge
configuration with Jz = 3 and thus a septet paired state. More- for both the B1 [Fig. 3(c)] and A1 [Fig. 3(d)] superconducting
over, since the interorbital pair amplitudes are not equivalent phases reveals an asymmetric line shape that occurs only for
for exchange of the orbital index, the resulting state includes the (100) termination. For the A1 the asymmetric profiles oc-
an orbital singlet configuration too. Such orbital singlet and cur for any choice of the d-vector structure (see Supplemental
quintet mixed parity state is a general mark of the pairing Material [43]). The asymmetric spectral weight distribution
structure at the edge for all the pairs of orbital channels in has a significant orbital dependence among the xz and yz chan-
the (zx, yz, xy) manifold. nels (see Supplemental Material [43] for details). This aspect
Asymmetric peaks in the density of states. We then move reflects the fact that xz and yz orbitals have a significant orbital
to the analysis of the density of the states at the edge of the anisotropy in the kinetic term along the (100) termination,

L020502-3
FUKAYA, YADA, TANAKA, GENTILE, AND CUOCO PHYSICAL REVIEW B 108, L020502 (2023)

FIG. 4. (a) Schematic illustration of two superconducting chains

in the xy plane with different spatial position along x axis and transla-
tional symmetry along y. tγi γ  with γ , γ  = a, b and i = x, y describe
the effective hopping amplitudes within the (xz, yz) manifold along
the x and y directions. (b) Local density of states for the (xz, yz)
y
orbitals as a function of the taa hopping parameter that controls the
orbital anisotropy, while keeping tbb y
= t and the strength of the C4
FIG. 3. Density of states (DOS) D(E ) in the bulk (a), (b) and rotation symmetry in the finite size two-chains problem. The reduc-
at the edge (c), (d) for a representative configuration with spin- y
tion of the taa amplitude yields a splitting of the orbital excitations
triplet/orbital-singlet/s-wave B1 (dx(xy,zx) , dy(xy,yz) ) = (|0 |, |0 |) and and an asymmetric profile of the density of states.
A1 states (dx(xy,zx) , dy(xy,yz) , dz(yz,zx) ) = (−|0 |, |0 |, 0.0). We choose a
representative orbital Rashba coupling as αOR /t = 0.30, while the
chemical potential is given by μ/t = 0.35. The other electronic singlet. This is instead generally realized due to the orbital
parameters are t  = 0.01t, t = 0.5t, λSO = 0.1t, and the gap am- Andreev reflections and the reduced crystalline symmetry at
plitude |0 | = 0.001t. the edge as demonstrated by the behavior of the anomalous
Green’s function in Fig. 2, even when in the bulk the s-wave
pairing channel is vanishing due to C4v symmetry constraints.
thus resulting in a substantial rotational symmetry breaking The model for the two chains includes orbital dependent elec-
at the boundary. On the other hand, for the (110) case, both tronic charge transfer along the x and y as it is expected to
xz and yz have a similar hopping amplitude along the (110) occur for the (xz, yz) states in the presence of rotation and
and the other diagonal (11̂0) direction with a low degree of mirror symmetry breaking (see Supplemental Material [43]
orbital anisotropy and rotational symmetry breaking. Since an for the details of the model Hamiltonian). We introduce the
asymmetric DOS is obtained for both the A1 and B1 phases, amplitudes tγi γ  with γ , γ  = a, b and i = x, y to describe the
we also observe that the nodal structure of the gap function is effective hopping within the (xz, yz) manifold along the x
not a needed feature for the achieved effects. and y directions. Due to the finite size of the model system,
Effective model and mechanism for asymmetric DOS. At assuming that taa = tbb and tab = 0, one recovers the rotational
this point, we would like to provide a minimal model to orbital symmetry and the density of states has the typical pro-
capture the mechanisms related to the occurrence of the asym- file of a conventional s-wave superconductor (bottom panel in
metric density of states at the edge. To this aim, we consider Fig. 4) with a full gap in the energy range [−|0 |, |0 |]. Then,
an effective model with equal spin (i.e., spin-triplet) pairing one can break the orbital rotational symmetry by unbalancing
and with only two-orbital degrees of freedom. We focus on the taa and tbb amplitudes. With this assumption the C4 rotation
the orbital manifold spanned by the (xz, yz) states which have symmetry is broken while the vertical mirror symmetry is
opposite parity with respect to the vertical mirrors (e.g., Mx ) preserved. For instance, one can consider the change of the
and are energy degenerate in the tetragonal configuration with hopping amplitude of the xz orbitals along the y directions.
C4v point group symmetry. This implies that at the edge they The hopping asymmetry acts like an effective k-dependent
get split and coupled by crystalline interactions that break mir- orbital field in the (xz, yz) manifold and thus it splits the peaks
ror and rotation symmetries. Then, in order to make a direct associated with the orbital excitations. The splitting and the
connection with the problem upon examination that breaks occurrence of spectral weight asymmetry can be observed in
translational symmetry and lacks rotation and vertical mirror the density of states [Fig. 4(b)] already for a small deviation of
y y
symmetries at the boundary, we consider a model Hamiltonian tbb from taa . The spectral weight profile gets more asymmetric
in real space for a limited number of chains. Here, on the basis as the amplitude of the orbital field increases by unbalancing
of the complete analysis of the anomalous correlator at the taa with respect to tbb . Similar modifications of the density
edge, we can assume that the pairing amplitude is given by of states can be obtained by varying the hopping ampli-
an s-wave equal spin configuration (spin triplet) with orbital tudes along the x direction. Along this line, by considering a

L020502-4
PARTICLE-HOLE SPECTRAL ASYMMETRY AT THE EDGE … PHYSICAL REVIEW B 108, L020502 (2023)

nonvanishing tab one can also introduce a term that breaks in the regime of gating across the Lifshitz transition when
both rotation and vertical mirror symmetries. As for the the (zx, yz) orbital manifold gets populated. We notice that
previous cases, such symmetry breaking yields effects on similar effects are also expected at the surface of doped topo-
the density of states that are similar to those reported in logical superconductors [74] assuming two-orbital odd-parity
Fig. 4. Additionally, one can also consider orbital triplet con- spin-triplet pairing [75]. Moreover, we argue that our findings
figurations in the orbital manifold and obtain particle-hole can be generalized to superconductors involving pseudospin
asymmetric spectra in the presence of rotation and mirror degrees of freedom akin to orbitals (e.g., sublattice, valley) for
symmetry breaking. terminations that yield a splitting of the pseudospin degrees of
The emerging physical scenario can be also qualita- freedom. This can be achieved, for instance, in bilayer Rashba
tively accounted by making the following analogy. Assuming superconductors [76] or in transition metal dichalcogenides
that the (xz, yz) states are configurations associated to [77].
a pseudospin 1/2, the achieved spin-triplet orbital-singlet Finally, we would like to discuss possible experimental
configuration in the presence of rotation and mirror sym- proposals in order to single out the asymmetric features of
metry breaking at the edge is analogous to a spin-singlet the density of states arising from the interorbital spin-triplet
configuration—spin up and down corresponding to the xz and pairing. In this context, the first aspect to consider is that
yz orbitals—subjected to a source of time reversal symmetry our findings indicate that the asymmetric profile arises in the
breaking with spin dependent mass in momentum space and presence of time-reversal symmetry. Hence, from an experi-
anisotropic magnetic field. Thus, taking into account this anal- mental perspective, in order to isolate the uncovered orbital
ogy, the crystalline interactions that break rotation and mirror dependent features, one has to avoid magnetic fields and
symmetries act as effective fields that break the time reversal preliminarily test that the normal state is not magnetic. In
in the pseudospin orbital space. principle, magnetic impurities can lead to Kondo resonances
Conclusions. We show the occurrence of asymmetric den- close to the Fermi level which are marked by an asymmetric
sity of states in time-reversal spin-triplet superconductors profile in the density of states. However, the energy scales
equipped with an orbital degree of freedom that at the edge of the Kondo features are typically larger than the supercon-
are subjected to a symmetry breaking field in the orbital ducting gap [14] and their evolution in the presence of an
space due to the reduced crystalline symmetry and orbital applied magnetic field can help to discern them from those
Andreev reflections. The achieved outcomes have several arising from the superconducting excitations. Instead, since
consequences concerning the materials and the detection of the orbital degrees and the reduced crystalline symmetry play
unconventional pairing. First, the asymmetry in the DOS can an important role in steering the asymmetric profile of the
be exploited to distinguish between acentric superconductors spectral function for multiorbital superconductors, we argue
with time-reversal spin-triplet pairs in a single-orbital odd- that the application of strain [78] or electric fields [79–81] can
parity channel (e.g., p-wave) with respect to those marked
be valuable means to investigate the occurrence of particle-
by a multiorbital even-parity (e.g., s-wave) structure. Second,
hole asymmetry in the density of states. This is because both
we have a clear-cut example of a superconductor equipped
strain and gate fields do not break the time-reversal symmetry
with Andreev bound states that results in asymmetric peaks
and one can implement suitable setups to control the strength
in the spectra that depend on the strength of interactions that
break the rotation and mirror symmetry. Such outcomes are of the dissipation effects. Having a profile of the density of
in principle applicable to the recently discovered anomalies states at the edge whose asymmetry is augmented or reduced
at the surface of the UTe2 superconductor. While the occur- by applying strain or gate fields may be an indirect evidence
rence of spin-triplet pairs in UTe2 has been experimentally of orbital dependent spin-triplet pairing.
tested through nuclear magnetic resonance probes, our results Another path to singling out the pairing induced particle-
indicate that the orbital degrees of freedom might play a role hole asymmetry is to exploit the tunability of the orbital
in setting out the structure of the spin-triplet pairing. This is population at the Fermi level. For instance, in multiorbital
supported by ARPES observations demonstrating that in the superconductors as LAO-STO heterostructures, the applica-
normal state, apart from heavy f bands, U 5d and Te p orbitals tion of gate voltage can tune the orbital occupation from a
contribute to the electronic structure at the Fermi level [55]. dominant xy character to a combination of xy and xz, yz orbital
Another application points to the class of 2D non- flavor. In this case, the occurrence of an asymmetric density
centrosymmetric oxides heterostructures, such as LaAlO3 / of states at the edge in concomitance with the changeover of
SrTiO3 (LAO-STO) [56,57], especially due to its remarkable the orbital population can be used to single out the orbital
orbital control of the superconducting critical temperature dependent spin-triplet pairing.
by electrostatic gating [58–60], together with Rashba spin- Finally, the potential to induce gate driven asymmetry
orbit coupling [61,62], the occupation of the Ti 3d orbitals in the density of states can be exploited in junctions based
(dxy , dzx , dyz ) [63,64], and the spin and orbital sources of Berry on orbital dependent spin-triplet pairing to achieve nonstan-
curvature [65,66]. The superconducting phase in LAO-STO dard magnetoelectric and topological effects, as compared
exhibits several anomalous properties [67–73] that cannot be to the case of single band p-wave spin-triplet superconduc-
easily addressed within a conventional single orbital spin- tors [82,83], as well as to uncoventional orbital Edelstein
singlet scenario. Our findings indicate that the presence of an effects [84].
even-parity orbital dependent spin-triplet pairing would lead
to gate tunable asymmetric spectral line shape at the edges M.C., P.G., and Y.F. acknowledge support by the project
or interface in STO based heterostructures. This is expected “Two-dimensional Oxides Platform for SPINorbitronics nan-

L020502-5
FUKAYA, YADA, TANAKA, GENTILE, AND CUOCO PHYSICAL REVIEW B 108, L020502 (2023)

otechnology (TOPSPIN)” funded by the MIUR-PRIN Bando This work was supported by JSPS KAKENHI Grants No.
2017, Grant No. 20177SL7HC. M.C. and Y.F. acknowledge JP20H00131, No. JP18H01176, and No. JP20H01857, and
support by the EU’s Horizon 2020 research and innovation the JSPS-EPSRC Core-to-Core program “Oxide Superspin”
program under Grant Agreement No. 964398 (SUPERGATE). international network (Grant No. JPJSCCA20170002).

[1] A. Andreev, The thermal conductivity of the intermediate state variational perspective, J. Phys.: Condens. Matter 19, 016217
in superconductors, Sov. Phys. JETP 19, 1228 (1964). (2007).
[2] C. Bruder, Andreev scattering in anisotropic superconductors, [18] C. Zou, Z. Hao, X. Luo, S. Ye, Q. Gao, M. Xu, X. Li, P. Cai,
Phys. Rev. B 41, 4017 (1990). C. Lin, X. Zhou, D.-H. Lee, and Y. Wang, Particle-hole asym-
[3] S. Kashiwaya and Y. Tanaka, Tunnelling effects on surface metric superconducting coherence peaks in overdoped cuprates,
bound states in unconventional superconductors, Rep. Prog. Nat. Phys. 18, 551 (2022).
Phys. 63, 1641 (2000). [19] T.-K. Ng and C. M. Varma, Experimental signatures of time-
[4] J. A. Sauls, Andreev bound states and their signatures, Philos. reversal-violating superconductors, Phys. Rev. B 70, 054514
Trans. R. Soc. A 376, 20180140 (2018). (2004).
[5] E. Prada, P. San-Jose, M. W. A. de Moor, A. Geresdi, E. J. H. [20] J. Linder, M. Cuoco, and A. Sudbø, Spin-active interfaces and
Lee, J. Klinovaja, D. Loss, J. Nygard, R. Aguado, and L. P. unconventional pairing in half-metal/superconductor junctions,
Kouwenhoven, From Andreev to Majorana bound states in hy- Phys. Rev. B 81, 174526 (2010).
brid superconductor-semiconductor nanowires, Nat. Rev. Phys. [21] M. Eschrig, Spin-polarized supercurrents for spintronics, Rep.
2, 575 (2020). Prog. Phys. 78, 104501 (2015).
[6] Y. Tanaka and S. Kashiwaya, Theory of Tunneling Spec- [22] B. Lu, P. Burset, K. Yada, and Y. Tanaka, Tunneling spec-
troscopy of d-Wave Superconductors, Phys. Rev. Lett. 74, 3451 troscopy and josephson current of superconductor-ferromagnet
(1995). hybrids on the surface of a 3d Ti, Supercond. Sci. Technol. 28,
[7] X.-L. Qi and S.-C. Zhang, Topological insulators and supercon- 105001 (2015).
ductors, Rev. Mod. Phys. 83, 1057 (2011). [23] B. Lu and Y. Tanaka, Study on Green’s function on topologi-
[8] M. Sato and Y. Ando, Topological superconductors: A review, cal insulator surface, Philos. Trans. R. Soc. A 376, 20150246
Rep. Prog. Phys. 80, 076501 (2017). (2018).
[9] K. Flensberg, F. von Oppen, and A. Stern, Engineered platforms [24] S. Ando, S. Ikegaya, S. Tamura, Y. Tanaka, and K. Yada,
for topological superconductivity and Majorana zero modes, Surface state of the interorbital pairing state in the Sr 2 RuO4
Nat. Rev. Mater. 6, 944 (2021). superconductor, Phys. Rev. B 106, 214520 (2022).
[10] C. Nayak, S. H. Simon, A. Stern, M. Freedman, and S. Das [25] F. Setiawan and J. D. Sau, Electron-boson-interaction induced
Sarma, Non-Abelian anyons and topological quantum compu- particle-hole symmetry breaking of conductance into subgap
tation, Rev. Mod. Phys. 80, 1083 (2008). states in superconductors, Phys. Rev. Res. 3, L032038 (2021).
[11] S. Ran, C. Eckberg, Q.-P. Ding, Y. Furukawa, T. Metz, S. R. [26] M. Cuoco and J. Ranninger, Superconductor-insulator tran-
Saha, I.-L. Liu, M. Zic, H. Kim, J. Paglione, and N. P. Butch, sition driven by local dephasing, Phys. Rev. B 70, 104509
Nearly ferromagnetic spin-triplet superconductivity, Science (2004).
365, 684 (2019). [27] M. Cuoco and J. Ranninger, From an insulating to a superfluid
[12] D. Aoki, A. Nakamura, F. Honda, D. Li, Y. Homma, Y. Shimizu, pair-bond liquid, Phys. Rev. B 74, 094511 (2006).
a. J. Sato, G. Knebel, J.-P. Brison, A. Pourret, D. Braithwaite, [28] B. D. Woods, J. Chen, S. M. Frolov, and T. D. Stanescu,
G. Lapertot, Q. Niu, M. Valiska, H. Harima, and J. Flouquet, Zero-energy pinning of topologically trivial bound states
Unconventional superconductivity in heavy fermion UTe2 , in multiband semiconductor-superconductor nanowires, Phys.
J. Phys. Soc. Jpn. 88, 043702 (2019). Rev. B 100, 125407 (2019).
[13] D. Aoki, J.-P. Brison, J. Flouquet, K. Ishida, Y. Tokunaga, and [29] C.-X. Liu, J. D. Sau, and S. Das Sarma, Role of dissipation in
Y. Yanase, Unconventional superconductivity in UTe2 , J. Phys.: realistic Majorana nanowires, Phys. Rev. B 95, 054502 (2017).
Condens. Matter 34, 243002 (2022). [30] S. R. Park, C. H. Kim, J. Yu, J. H. Han, and C. Kim, Orbital-
[14] L. Jiao, S. Howard, S. Ran, Z. Wang, J. O. Rodriguez, M. Angular-Momentum Based Origin of Rashba-Type Surface
Sigrist, Z. Wang, N. P. Butch, and V. Madhavan, Chiral su- Band Splitting, Phys. Rev. Lett. 107, 156803 (2011).
perconductivity in heavy-fermion metal UTe2 , Nature (London) [31] J.-H. Park, C. H. Kim, J.-W. Rhim, and J. H. Han, Orbital
579, 523 (2020). Rashba effect and its detection by circular dichroism angle-
[15] Y. Tanuma, Y. Tanaka, M. Yamashiro, and S. Kashiwaya, Local resolved photoemission spectroscopy, Phys. Rev. B 85, 195401
density of states on rough surfaces of dx2 −y2 -wave superconduc- (2012).
tors, Phys. Rev. B 57, 7997 (1998). [32] G. Khalsa, B. Lee, and A. H. MacDonald, Theory of t2g
[16] Y. Tanuma, Y. Tanaka, M. Ogata, and S. Kashiwaya, Quasipar- electron-gas Rashba interactions, Phys. Rev. B 88, 041302(R)
ticle states near surfaces of high−T c superconductors based on (2013).
the extended t − j model, Phys. Rev. B 60, 9817 (1999). [33] P. Kim, K. T. Kang, G. Go, and J. H. Han, Nature of orbital
[17] H.-Y. Yang, F. Yang, Y.-J. Jiang, and T. Li, On the origin of and spin Rashba coupling in the surface bands of SrTiO3 and
the tunnelling asymmetry in the cuprate superconductors: A KTaO3 , Phys. Rev. B 90, 205423 (2014).

L020502-6
PARTICLE-HOLE SPECTRAL ASYMMETRY AT THE EDGE … PHYSICAL REVIEW B 108, L020502 (2023)

[34] M. T. Mercaldo, P. Solinas, F. Giazotto, and M. Cuoco, Elec- [51] J. Linder and A. V. Balatsky, Odd-frequency superconductivity,
trically Tunable Superconductivity through Surface Orbital Rev. Mod. Phys. 91, 045005 (2019).
Polarization, Phys. Rev. Appl. 14, 034041 (2020). [52] A. M. Black-Schaffer and A. V. Balatsky, Odd-frequency super-
[35] Y. Fukaya, S. Tamura, K. Yada, Y. Tanaka, P. Gentile, and M. conducting pairing in multiband superconductors, Phys. Rev. B
Cuoco, Interorbital topological superconductivity in spin-orbit 88, 104514 (2013).
coupled superconductors with inversion symmetry breaking, [53] Y. Asano and A. Sasaki, Odd-frequency Cooper pairs in two-
Phys. Rev. B 97, 174522 (2018). band superconductors and their magnetic response, Phys. Rev.
[36] Y. Fukaya, S. Tamura, K. Yada, Y. Tanaka, P. Gentile, and M. B 92, 224508 (2015).
Cuoco, Spin-orbital hallmarks of unconventional superconduc- [54] H. Ebisu, B. Lu, J. Klinovaja, and Y. Tanaka, Theory of time-
tors without inversion symmetry, Phys. Rev. B 100, 104524 reversal topological superconductivity in double Rashba wires:
(2019). Symmetries of Cooper pairs and Andreev bound states, Prog.
[37] Y. Fukaya, K. Yada, Y. Tanaka, P. Gentile, and M. Cuoco, Theor. Exp. Phys. 2016, 083101 (2016).
Orbital tunable 0 − π transitions in Josephson junctions with [55] L. Miao, S. Liu, Y. Xu, E. C. Kotta, C.-J. Kang, S. Ran, J.
noncentrosymmetric topological superconductors, Phys. Rev. B Paglione, G. Kotliar, N. P. Butch, J. D. Denlinger, and L. A.
102, 144512 (2020). Wray, Low Energy Band Structure and Symmetries of UTe2
[38] M. Sato, Y. Tanaka, K. Yada, and T. Yokoyama, Topology of from Angle-Resolved Photoemission Spectroscopy, Phys. Rev.
Andreev bound states with flat dispersion, Phys. Rev. B 83, Lett. 124, 076401 (2020).
224511 (2011). [56] A. Ohtomo and H. Y. Hwang, A high-mobility electron gas at
[39] K. Yada, M. Sato, Y. Tanaka, and T. Yokoyama, Surface density the LaAlO3 /SrTiO3 heterointerface, Nature (London) 427, 423
of states and topological edge states in noncentrosymmetric (2004).
superconductors, Phys. Rev. B 83, 064505 (2011). [57] N. Reyren, S. Thiel, A. D. Caviglia, L. F. Kourkoutis, G.
[40] P. M. R. Brydon, A. P. Schnyder, and C. Timm, Topologically Hammerl, C. Richter, C. Schneider, T. Kopp, A.-S. Ruetschi,
protected flat zero-energy surface bands in noncentrosymmetric D. Jaccard, M. Gabay, D. A. Muller, J.-M. M. Triscone, J.
superconductors, Phys. Rev. B 84, 020501(R) (2011). Mannhart, L. F. Kourkoutis, and G. Hammerl, Superconduct-
[41] M. T. Mercaldo, M. Cuoco, and P. Kotetes, Magnetic-field- ing interfaces between insulating oxides, Science 317, 1196
induced topological reorganization of a p-wave superconductor, (2007).
Phys. Rev. B 94, 140503(R) (2016). [58] A. D. Caviglia, S. Gariglio, N. Reyren, D. Jaccard, T. Schneider,
[42] Y. Fukaya, Y. Tanaka, P. Gentile, K. Yada, and M. Cuoco, M. Gabay, S. Thiel, G. Hammerl, J. Mannhart, and J.-M.
Anomalous Josephson coupling and high-harmonics in non- Triscone, Electric field control of the LaAlO3 /SrTiO3 interface
centrosymmetric superconductors with s-wave spin-triplet pair- ground state, Nature (London) 456, 624 (2008).
ing, npj Quantum Mater. 7, 99 (2022). [59] H. Thierschmann, E. Mulazimoglu, N. Manca, S. Goswami,
[43] See Supplemental Material at http://link.aps.org/supplemental/ T. M. Klapwijk, and A. Caviglia, Transport regimes of a
10.1103/PhysRevB.108.L020502 for the methodology to obtain split gate superconducting quantum point contact in the two-
the boundary Green’s function. Then, we consider an effective dimensional LaAlO3 /SrTiO3 superfluid, Nat. Commun. 9, 2276
two-orbital model with a pairing structure having orbital singlet (2018).
symmetry and crystalline interactions that break the rotational [60] S. Hurand, A. Jouan, C. Feuillet-Palma, G. Singh, J. Biscaras,
symmetry and evaluate the density of states at the edge. We E. Lesne, N. Reyren, A. Barthélémy, M. Bibes, J. E. Villegas,
show that symmetric surface density of states generally occurs C. Ulysse, X. Lafosse, M. Pannetier-Lecoeur, S. Caprara, M.
in a single-orbital noncentrosymmetric superconductor with Grilli, J. Lesueur, and N. Bergeal, Field-effect control of su-
Rashba coupling. perconductivity and Rashba spin-orbit coupling in top-gated
[44] P. A. Lee and D. S. Fisher, Anderson Localization in Two LaAlO3 /SrTiO3 devices, Sci. Rep. 5, 12751 (2015).
Dimensions, Phys. Rev. Lett. 47, 882 (1981). [61] A. D. Caviglia, M. Gabay, S. Gariglio, N. Reyren, C.
[45] A. Umerski, Closed-form solutions to surface Green’s func- Cancellieri, and J.-M. Triscone, Tunable Rashba Spin-Orbit
tions, Phys. Rev. B 55, 5266 (1997). Interaction at Oxide Interfaces, Phys. Rev. Lett. 104, 126803
[46] K. Kawai, K. Yada, Y. Tanaka, Y. Asano, A. A. Golubov, and (2010).
S. Kashiwaya, Josephson effect in a multiorbital model for [62] M. Ben Shalom, M. Sachs, D. Rakhmilevitch, A. Palevski, and
Sr 2 RuO4 , Phys. Rev. B 95, 174518 (2017). Y. Dagan, Tuning Spin-Orbit Coupling and Superconductivity
[47] Y. Tanaka, M. Sato, and N. Nagaosa, Symmetry and topol- at the SrTiO3 /LaAlO3 Interface: A Magnetotransport Study,
ogy in superconductors -odd-frequency pairing and edge states, Phys. Rev. Lett. 104, 126802 (2010).
J. Phys. Soc. Jpn. 81, 011013 (2012). [63] A. Joshua, S. Pecker, J. Ruhman, E. Altman, and S. Ilani, A
[48] Y. Tanaka and A. A. Golubov, Theory of the Proximity Effect universal critical density underlying the physics of electrons at
in Junctions with Unconventional Superconductors, Phys. Rev. the LaAlO3 /SrTiO3 interface, Nat. Commun. 3, 1129 (2012).
Lett. 98, 037003 (2007). [64] G. Herranz, G. Singh, N. Bergeal, A. Jouan, J. Lesueur, J.
[49] Y. Tanaka, Y. Tanuma, and A. A. Golubov, Odd-frequency Gázquez, M. Varela, M. Scigaj, N. Dix, F. Sánchez, and J.
pairing in normal-metal/superconductor junctions, Phys. Rev. Fontcuberta, Engineering two-dimensional superconductivity
B 76, 054522 (2007). and rashba spin-orbit coupling in LaAlO3 /SrTiO3 quantum
[50] M. Eschrig, T. Lofwander, T. Champel, J. C. Cuevas, J. Kopu, wells by selective orbital occupancy, Nat. Commun. 6, 6028
and G. Schon, Symmetries of pairing correlations in supercon- (2015).
ductor ferromagnet nanostructures, J. Low Temp. Phys. 147, [65] E. Lesne, Y. G. Saǧlam, R. Battilomo, M. T. Mercaldo, T. C.
457 (2007). van Thiel, U. Filippozzi, C. Noce, M. Cuoco, G. A. Steele, C.

L020502-7
FUKAYA, YADA, TANAKA, GENTILE, AND CUOCO PHYSICAL REVIEW B 108, L020502 (2023)

Ortix, A. D. Caviglia, Designing spin and orbital sources of [74] T. Mizushima, A. Yamakage, M. Sato, and Y. Tanaka, Dirac-
Berry curvature at oxide interfaces, Nat. Mater. 22, 576 (2023). fermion-induced parity mixing in superconducting topological
[66] M. T. Mercaldo, C. Noce, A. D. Caviglia, M. Cuoco, C. insulators, Phys. Rev. B 90, 184516 (2014).
Ortix, Orbital design of Berry curvature: Pinch points and giant [75] L. Fu and E. Berg, Odd-Parity Topological Superconductors:
dipoles induced by crystal fields, npj Quantum Mater. 8, 12 Theory and Application to Cux Bi2 Se3 , Phys. Rev. Lett. 105,
(2023). 097001 (2010).
[67] G. Singh, A. Jouan, G. Herranz, M. Scigaj, F. Sánchez, L. [76] S. Nakosai, Y. Tanaka, and N. Nagaosa, Topological Super-
Benfatto, S. Caprara, M. Grilli, G. Saiz, F. Couedo, C. Feuillet- conductivity in Bilayer Rashba System, Phys. Rev. Lett. 108,
Palma, J. Lesueur, and N. Bergeal, Gap suppression at a Lifshitz 147003 (2012).
transition in a multi-condensate superconductor, Nat. Mater. 18, [77] X. Xu, W. Yao, D. Xiao, and T. F. Heinz, Spin and pseudospins
948 (2019). in layered transition metal dichalcogenides, Nat. Phys. 10, 343
[68] T. V. Trevisan, M. Schütt, and R. M. Fernandes, Uncon- (2014).
ventional Multiband Superconductivity in Bulk SrTiO3 and [78] J. Mutch, W.-C. Chen, P. Went, T. Qian, I. Z. Wilson, A.
LaAlO3 /SrTiO3 interfaces, Phys. Rev. Lett. 121, 127002 Andreev, C.-C. Chen, and J.-H. Chu, Evidence for a strain-
(2018). tuned topological phase transition in ZrTe5 , Sci. Adv. 5,
[69] V. V. Bal, M. M. Mehta, S. Ryu, H. Lee, C. M. Folkman, eaav9771 (2019).
C. B. Eom, and V. Chandrasekhar, Gate-tunable superconduct- [79] G. De Simoni, F. Paolucci, P. Solinas, E. Strambini, F. Giazotto,
ing weak link behavior in top-gated LaAlO3 -SrTiO3 , Appl. Metallic supercurrent field-effect transistor, Nat. Nanotechnol.
Phys. Lett. 106, 212601 (2015). 13, 802 (2018).
[70] D. Stornaiuolo, D. Massarotti, R. Di Capua, P. Lucignano, G. P. [80] G. De Simoni, S. Battisti, N. Ligato, M. T. Mercaldo, M.
Pepe, M. Salluzzo, and F. Tafuri, Signatures of unconventional Cuoco, and F. Giazotto, Gate control of the currentflux relation
superconductivity in the LaAlO3 /SrTiO3 two-dimensional sys- of a Josephson quantum interferometer based on proximitized
tem, Phys. Rev. B 95, 140502(R) (2017). metallic nanojuntions, ACS Appl. Electron. Mater. 3, 3927
[71] A. Kalaboukhov, P. P. Aurino, L. Galletti, T. Bauch, F. (2021).
Lombardi, D. Winkler, T. Claeson, and D. Golubev, Homo- [81] M. T. Mercaldo, F. Giazotto, and M. Cuoco, Spectroscopic sig-
geneous superconductivity at the LaAlO3 /SrTiO3 interface natures of gate-controlled superconducting phases, Phys. Rev.
probed by nanoscale transport, Phys. Rev. B 96, 184525 (2017). Res. 3, 043042 (2021).
[72] L. Kuerten, C. Richter, N. Mohanta, T. Kopp, A. Kampf, [82] M. T. Mercaldo, P. Kotetes, and M. Cuoco, Magnetoelectrically
J. Mannhart, and H. Boschker, In-gap states in supercon- tunable Andreev bound state spectra and spin polarization in
ducting LaAlO3 − SrTiO3 interfaces observed by tunneling p-wave Josephson junctions, Phys. Rev. B 100, 104519 (2019).
spectroscopy, Phys. Rev. B 96, 014513 (2017). [83] M. T. Mercaldo, P. Kotetes, and M. Cuoco, Magnetic manipu-
[73] G. Singh, C. Guarcello, E. Lesne, D. Winkler, T. Claeson, T. lation of topological states in p-wave superconductors, Phys. B:
Bauch, F. Lombardi, A. D. Caviglia, R. Citro, M. Cuoco, and Condens. Matter 536, 730 (2018).
A. Kalaboukhov, Gate-tunable pairing channels in supercon- [84] L. Chirolli, M. T. Mercaldo, C. Guarcello, F. Giazotto, and M.
ducting non-centrosymmetric oxides nanowires, npj Quantum Cuoco, Colossal Orbital Edelstein Effect in Noncentrosymmet-
Mater. 7, 2 (2022). ric Superconductors, Phys. Rev. Lett. 128, 217703 (2022).

L020502-8