Transmission through rectangular potentials in semimetals featuring quadratic dispersion

Ipsita Mandal∗
Department of Physics, Shiv Nadar Institution of Eminence (SNIoE),
Gautam Buddha Nagar, Uttar Pradesh 201314, India

We revisit the problem of transmission of quasiparticles through a rectangular potential barrier

for semimetals featuring quadratic-in-momentum band-crossings at a nodal point. Although this
was considered in Annals of Physics 419 (2020) 168235, the solutions corresponding to evanescent
waves were missed, leading to a partial fulfillment of the boundary conditions required to determine
the piecewise-continuous wavefunctions. In this paper, our aim is to correct those shortcomings,
recompute the transmission coefficients, and show the resulting behaviour of the conductivity and
the Fano factor for some representative parameter values.

Contents

I. Introduction 1

II. 2d QBCP 2
A. Formalism 4
B. Transmission coefficients, conductivity, and Fano factors 8

III. 3d QBCP 8
A. Formalism 9
B. Transmission coefficients, conductivity, and Fano factors 12

IV. Summary and outlook 13

References 14

I. Introduction

Nodal-point semimetals are materials which feature twofold or multifold band-crossing points in the Brillouin zone
(BZ) where two or more bands cross. This implies that when the chemical potential cuts the nodal point, the Fermi
surface shrinks to a Fermi node. The poster child examples of such semimetals comprise (1) graphene [1], for the case
of two-dimensional (2d) systems; and (2) Dirac and Weyl semimetals [2]. For each of these cases, the bands crossing at
a node show linear-in-momentum dependence. However, there exist 2s and 3d semimetals where the bands crossing at
a node show quadratic-in-momentum dependence [3–18], which are sometimes dubbed as quadratic-band-crossing points
(QBCPs). While 2d QBCPs are realised in checkerboard [3] (at half-filling), Kagome [3] (at one-third-filling), and Lieb
[4] lattices, various pyrochlore iridates A2 Ir2 O7 (where A stands for a lanthanide element [19, 20]) host 3d QBCPs. Such
bandstructures have also been realised in 3d gapless semiconductors in the presence of a sufficiently strong spin-orbit
coupling [21], such that the resulting model is of relevance for materials like gray tin (α-Sn) and mercury telluride (HgTe).
These systems are also known as “Luttinger semimetals” due to the fact that the low-energy quasipartcles (in the vicinity
of a node) are captured by the Luttinger Hamiltonian of inverted band-gap semiconductors [6, 22–24]. In this paper, we
will compute the transmission and reflection coefficients of the quasiparticles in the vicinity of 2d and 3d QBCPs, when
moving across a rectangular potential, which is chosen to lie in a path along the x-axis or z-axis. The set-up is represented
schematically in Fig. 1. Although this was considered in Ref. [11], the solutions corresponding to evanescent waves were
missed (see Refs. [25–28] for analogous situations in other kinds of semimetals), leading to a partial fulfillment of the
boundary conditions required to determine the piecewise-continuous wavefunctions. With the correct solutions in place,
we will determine the transport-characteristics reflected in the reflection and transmission coefficients. After determining

∗ [Link]@[Link]
 2

kx
V0
E
L

FIG. 1. Schematics of transmission through a potential barrier in a system hosting a QBCP. The upper panel shows the schematic
diagrams of the spectrum of quasiparticles about a QBCP, when a potential barrier is encountered while mocing along the x-direction.
The lower panel represents the schematic diagram of the transport across the potential barrier. The Fermi level (indicated by dotted
lines) lies in the conduction band outside the barrier, and in the valence band inside it. The blue-coloured areas of the bands indicate
occupied states.

the transmission amplitude (T ), we will compute the conductance (G), given by the Landauer formula of [29–31]
e2 X
G= Tn . (1)
2πℏ n

Here, e is the charge of one electron and n labels the transverse momentum modes in the the strip of the material (in
the experimental set-up). When an external potential difference Φ is applied across a circuit, shot noise is the physical
quantity which provides a measure of the fluctuations of the electric current-density away from its average value. At
3
zero temperature, while the actual shot noise is defined by S = e π|Φ|
P
ℏ n Tn (1 − Tn ) , the Poisson noise is given by the
3
e |Φ| P
expression SP = π ℏ n Tn [31]. The Poisson noise is the value of the noise that would be measured if the system
produced noise due to quasiparticles carrying a single set of the relevant quantum numbers. A convenient measure of the
sub-Poissonian shot noise is the Fano factor (F ), which is the ratio defined as [31]
P
S Tn (1 − Tn )
F = = nP . (2)
SP ñ Tñ

The paper is organized as follows. Sec. II deals with the 2d QBCPs and Sec. III deals with 3d QBCPs. The nature
of conductivity and Fano factors are compared with those of other types of bandstructures. Finally, we conclude with a
summary and outlook in Sec. IV. In all our expressions, we will be using the natural units, which means that the reduced
Planck’s constant (ℏ), the speed of light (c), and the magnitude of a single electronic charge (e) are each set to unity.

II. 2d QBCP

A QBCP in a 2d system harbours a twofold band-crossing, represented by the low-energy effective Hamiltonian [3]
kin 1 h
i
H2d (kx , ky ) = 2 kx ky σx + ky2 − kx2 σz (3)
2m
in the momentum space. The energy eigenvalues are captured by ± ε2d (kx , ky ), where
kx2 + ky2
ε2d (kx , ky ) = , (4)
2m
 3

FIG. 2. 2d QBCP: The polar plots show the transmission coefficient, T (E, V0 , ϕ) E<V (upper panel), and the reflection coefficient
0
R(E, V0 , ϕ) E≤V (lower panel), as functions of the incident angle ϕ, for some representative values of 2 m L2 V0 . The values of E
0
are shown in the plot-legends.

where the “+” and “−” signs, in the usual conventions, refer to the positive-energy (or conduction) and negative-energy
(or valence) bands, respectively. The corresponding orthonormal eigenvectors are given by
1 1
ΨT+ = q (ky kx ) and ΨT− = q (− kx ky ) , (5)
kx2 + ky2 kx2 + ky2

respectively.
The 2d system is modulated by a rectangular electric potential barrier of height V0 and width L [cf. Fig. 1], giving rise
to an x-dependent potential energy function of the form
(
V0 for 0 ≤ x ≤ L
V (x) = . (6)
0 otherwise

Hence, we need to consider the total Hamiltonian,

tot kin
H2d = H2d (− i ∂x , − i ∂y ) + V (x) , (7)
 4

FIG. 3. 2d QBCP: The polar plots show the transmission coefficient, T (E, V0 , ϕ) E>V (upper panel), and the reflection coefficient
0
T (E, V0 , ϕ) E>V (lower panel), as functions of the incident angle ϕ, for some representative values of 2 m L2 V0 . The values of E
0
are shown in the plot-legends.

in the position space. As mentioned earlier, here, the transport direction is chosen along the x-axis. We fix the Fermi-
energy level at a value of E > 0 in the region outside the potential barrier, which can be adjusted in an experimental
set-up by either chemical doping or an external gate-voltage.

A. Formalism

For a material of a sufficiently large transverse dimension W , the boundary conditions should be irrelevant for the bulk
response, and we use this freedom to simplify the calculation. Here, on a physical wavefunciton Ψtot we impose periodic
boundary conditions of Ψtot (x, W ) = Ψtot (x, 0) . The transverse momentum ky is conserved, and it is quantized√due to
the periodicity in the transverse width W and, hence, takes the form of ky = πWn ≡ qn , where n ∈ Z and |ky | ≤ 2 m E.
For the longitudinal direction, we seek plane wave solutions of the form ei kx x . Then the full wavefunction is given by:

Ψtot (x, y, qn ) = const. × Ψn (x) ei qn y , (8)

 5

FIG. 4. 2d electron-gas: The polar plots show the transmission coefficient, T (E, V0 , ϕ) E<V (upper panel), and the reflection
0
coefficient R(E, V0 , ϕ) E≤V (lower panel), as functions of the incident angle ϕ, for some representative values of 2 m L2 V0 . The
0
values of E are shown in the plot-legends.

For any mode of given transverse (to the propagation direction) momentum component qn (with |qn | ≤ 2 m E), we can
determine the x-component of the wavevectors of the incoming, reflected, and transmitted waves, by solving 4 m2 E 2 =

2
kx2 + qn2 ⇒ kx2 = ± 2 m E −qn2 . In fact, in the regions x < 0 and x > L, this relation leads to the following four solutions:
p p
kx = ± 2 m E − qn2 and kx = ± i 2 m E + qn2 . (9)
Consequently, in addition to the propagating plane-wave solutions, there are also evanescent
p waves present [25–28],
p charac-
terized by complex values.1 For x < 0, we need to consider the solutions kx = − 2 m E − qn2 and kx = − i 2 m E + qn2
while determining the overall
p reflected wave, which p clearly comprises evanescent parts. For x > L, we need to con-
sider the solutions kx = 2 m E − qn2 and kx = i 2 m E + qn2 while determining the overall transmitted wave, which
again comprises
p evanescent parts. Insidepthe potential-barrier region, the x-components of the momentum are given by
kx = ± 2 m |E − V0 | − qn2 and kx = ± i 2 m |E − V0 | + qn2 . This situation resembles the case of bilayer graphene [32].

1 In general, if the Fermi energy cuts the bands at an energy µ, for propagation along the x-direction, the corresponding propagating“right-
moving” plane waves will have the factor ei sgn(µ) kx x . This just implies that if the propagating quasiparticles are occupying the upper
(lower) band, then they have a positive (negative) group velocity.
 6

FIG. 5. 2d electron-gas: The polar plots show the transmission coefficient, T (E, V0 , ϕ) E>V (upper panel), and the reflection
0
coefficient T (E, V0 , ϕ) E>V (lower panel), as functions of the incident angle ϕ, for some representative values of 2 m L2 V0 . The
0
values of E are shown in the plot-legends.

We follow usual procedure of matching the piecewise-continuous regions (see, e.g., Refs. [33, 34]) to compute the
reflection and transport coefficients. Here, we consider the transport of positive energy states (i.e., Ψ+ ) corresponding to
electron-like particles. The transport of hole-like excitations (i.e., Ψ− ) will be similar. Hence, the Fermi level outside the
potential barrier is adjusted to the value E (with E > 0). A scattering state ∼ Ψn (x) ei qn y , labelled by the index n, is
constructed in a piecewise fashion as


ϕL (x)
 for x < 0
Ψn (x) = ϕM (x) for 0 ≤ x ≤ L , (10)

ϕ (x)
R for x > L
 7

(a)

(b)

FIG. 6. Plots of the conductivity (σ) and Fano factor (F ), as functions of E/V0 , for various values of 2 m L2 V0 . The subfigures (a)
and (b) correspond to 2d QBCP and 2d electron-gas, respectively. Here, σ and F are unitless.

where
Ψ+ (kin , qn ) ei kin x + rn Ψ+ (− kin , qn ) e−i kin x
ϕL (x) = p + r̃n Ψ− (− i κ, qn ) e−κ |x| ,
V(kin , qn )
ϕM (x)
h i
= αn Ψ+ (kmid , qn ) ei kmid x + βn Ψ+ (− kmid , qn ) e−i kmid x + α̃n Ψ− (i κmid , qn ) e− κmid x + β̃n Ψ− (− i κmid , qn ) eκmid x
× Θ (E − V0 )
h i
+ αn Ψ− (kmid , qn ) ei kmid x + βn Ψ− (− kmid , qn ) e−i kmid x + α̃n Ψ+ (i κmid , qn ) e−κmid x + β̃n Ψ+ (− i κmid , qn ) eκmid x
× Θ (V0 − E) ,
ei kin (x−L)
ϕR (x) = tn Ψ+ (kin , qn ) p + t̃n Ψ− (i κ, qn ) e− κ x ,
V(kin , qn )
p kin p
kin = 2 m E − qn2 , V(kin , qn ) ≡ |∂kin ε2d (kin , qn )| = , κ= 2 m E + qn2 ,
p p m
kmid = 2 m |E − V0 | − qn2 , κmid = 2 m |E − V0 | + qn2 . (11)

Here, the magnitude of the group-velocity, captured by V(kin , qn ), is needed to define the scattering matrix comprising
the transmission and reflection coefficients. It is important to note that, for V0 > E, the Fermi level within the potential
barrier lies within the valence band and we must use the valence band wavefunctions in that region — this input is
captured by the usage of the Heaviside-theta functions.
tot tot
The boundary conditions can be obtained by integrating the equation H2d Ψ = E Ψtot over a small interval in the
 8

x-direction, around the points x = 0 and x = L, in two consecutive steps. The results are obtained as the two components
of the wavefunction and its x-derivative being continuous at the boundaries, as shown below:

ϕL (0) = ϕM (0) and ∂x ϕL (x) x=0

= ∂x ϕM (x) x=0
;
ϕM (L) = ϕR (L) and ∂x ϕM (x) x=L
= ∂x ϕL (x) x=L
. (12)

These conditions are sufficient to guarantee the continuity of the flux of the probability-current density along the x-
direction. From the matching of the wavefunction and its derivatives, we have 2 × 2 = 4 matrix-equations from the
two boundaries. For 2d QPCBs, each of these matrix-equations can be separated into two components, since each
wavevector has two components. Therefore, we have 4 × 2 = 8 equations for the eight undetermined coefficients
{rn , r̃n , αn , βn , α̃n , β̃n , tn , t̃n }. The explicit analytical expressions for t and r are extremely long and, hence, we re-
frain from showing them here. Instead, we represent their characteristics via Figs. 2 and 3.
We would like to alert the reader that the evanescent-wave solutions do not contribute to the reflection and transmission
coefficients, as they decay off as we move away from the barrier-junctions. Hence, the transmission and reflection coefficients
at an energy E are given by

T (E, V0 , ϕ) = |tn (E, V0 )|2 and R(E, V0 , ϕ) = |rn (E, V0 )|2 , (13)

respectively, where ϕ = tan−1 (qn /kin ) is the incident angle of the incoming plane-wave. Figs. 2 and 3 illustrate the polar
plots of T (E, V0 , ϕ) and R(E, V0 , ϕ), as functions of ϕ, for some representative values of V0 , L, and E.
Let us compare the results obtained for a 2d QBCP with those in a 2d electron-gas. For normal metals, we have only
one electronic band to consider, which the Fermi energy will intersect (irrespective of the height of the barrier). Using the
continuity of the wavefunctions and their x-derivatives at the two ends of the barrier, we can easily find the transmission
amplitude to be given by

2 i kmid kin
for E ≤ V0
tn (E, V0 ) = ( mid in )
 2
k + k2 sin(kmid L)+2 i kmid kin cos(kmid L)
− 2 κmid kin . (14)
i (kin −κmid ) sinh(κmid L)−2 κmid kin cosh(κmid L)
 2 2 for E > V0

Figs. 4 and 5 illustrate the polar plots of the transmission and reflection coefficients to provide a comparison with the case
of the 2d QBCP.

B. Transmission coefficients, conductivity, and Fano factors

√ Let us assume W√to be very large such that √ qn can effectively be treated as a continuous variable. Using kin =
2 m E cos ϕ, ky = 2 m E sin ϕ, and dky = 2 m E | cos ϕ| dϕ (using the continuum limit when W ≫ L), we consider
the zero-temperature limit and an applied voltage of very small magnitude. Remembering that ℏ = 1 in the natural units,
the conductance is given by [cf. Eq. (1)]

e2 X e2 X  π
G(E, V0 ) = |tn |2 = |tn |2 (qn − qn−1 ) where ∆q =
2π n 2 π ∆q n W
√
2mE
√
e2 W e2 W 2 m E π/2
Z Z
→ dk y T (E, V 0 , ky ) = dϕ cos ϕ T (E, V0 , ϕ) . (15)
2 π 2 − √2 m E 2 π2 −π/2

We note that the conductivity is given by [34]

√
e2 L 2 m E π/2
Z
L
σ(E, V0 ) = G(E, V0 ) = dϕ cos ϕ T (E, V0 , ϕ) . (16)
W 2 π2 −π/2

In Fig. 6(a), we illustrate the conductivity and the Fano factor [cf. Eq. (2)], as functions of E/V0 , for some values
of 2 m L2 V0 . Side by side, the corresponding curves [generated using Eq. (14)] corresponding for a 2d electron-gas are
provided in Fig. 6(b) for the sake of comparison.

III. 3d QBCP

A QBCP in a 3d system harbours a fourfold band-crossing, where the four bands form a four-dimensional representation
of the underlying lattice-symmetry group [8]. Then the standard (k · p) Hamiltonian for the particle-hole symmetric system
 9

can be written by using a representation of the five 4 × 4 Euclidean Dirac matrices Γa as [23, 35]
5
kin ℏ2 X
H3d (kx , ky , kz ) = da (k) Γa . (17)
2 m a=1

The Γa -matrices form one of the (two possible) irreducible four-dimensional Hermitian representations of the five-
component Clifford algebra, defined by the anticommutator { Γa , Γb } = 2 δab . The five anticommuting gamma-matrices
can always be chosen such that three are real and two are imaginary [23, 36]. In the representation used here, (Γ1 , Γ3 , Γ5 )
are real and (Γ2 , Γ4 ) are imaginary:

Γ1 = σ1 ⊗ σ0 , Γ2 = σ2 ⊗ σ0 , Γ3 = σ3 ⊗ σ1 , Γ4 = σ3 ⊗ σ2 , Γ5 = σ3 ⊗ σ3 . (18)

The five functions da (k) are the real ℓ = 2 spherical harmonics, with the following structure:
√
√ √ √ 3 2 1
(kx − ky2 ) , d5 (k) = 2 kz2 − kx2 − ky2 .


d1 (k) = 3 ky kz , d2 (k) = 3 kx kz , d3 (k) = 3 kx ky , d4 (k) = (19)
2 2
The energy eigenvalues are given by ± ε3d (kx , ky , kz ), where


ℏ2 kx2 + ky2 + kz2
ε3d (kx , ky , kz ) = . (20)
2m
The “+” and “−” signs, as usual, refer to the conduction and valence bands. Each of these bands is doubly degenerate.
A set of orthonormal eigenvectors is given by the following:


T 1  (kx + i ky ) (k + kz ) i (k + 3 kz ) i −2 kz (k + kz ) + kx2 + ky2 
Ψ+,1 = − √ − √ 1 ,
N+,1 (kx − i ky )2 3 (kx − i ky ) 3 (kx − i ky )2

!
T 1 (kx + i ky ) (k − kz ) i (k − 3 kz ) i 2 kz (k − kz ) + kx2 + ky2
Ψ+,2 = −√ − √ 1 ,
N+,2 (kx − i ky )2 3 (kx − i ky ) 3 (kx − i ky )2

!
T 1 i (k + kz ) k − kz i 2 kz (kz − k) + kx2 + ky2
Ψ−,1 = −√ 1 − √ ,
N−,1 3 (kx − i ky ) kx + i ky 3 (kx + i ky )2
2 2

!
1 i (k − k z ) k + k z i 2 k z (k + kz ) + k x + k y
ΨT−,2 = √ − 1 − √ , (21)
N−,2 3 (kx − i ky ) kx + i ky 3 (kx + i ky )2
q
where k = kx2 + ky2 + kz2 . The “+” (“−”) subscript indicates an eigenvector corresponding to the positive (negative)
1 1
eigenvalue. The symbols N±,1 and N±,2 denote the corresponding normalization factors.
The 3d system is modulated by a square electric potential barrier of height V0 and width L. Here, we choose the
z-axis as the transport direction, and place the chemical potential at an energy E > 0 in the region outside the potential
barrier. Since the system is isotropic, it does not matter which axis we choose. But it is just a matter of convenience
while transforming to the conventional spherical polar coordinates. Therefore, we have
(
V0 for 0 ≤ z ≤ L
V (z) = . (22)
0 otherwise

As before, we need to consider the total Hamiltonian in the position space as

tot kin
H3d = H3d (−i ∂x , −i ∂y , −i ∂z ) + V (z) . (23)

All this amounts to replacing kx by kz in Fig. 1.

A. Formalism

We consider the tunneling in a slab of height and width W . Again, we assume that the material has a sufficiently large
width W along each of the two transverse directions, such that the boundary conditions are irrelevant for the bulk response,
and impose the following periodic boundary conditions: Ψ̃tot (x, 0, z) = Ψ̃tot (x, W, z) , Ψ̃tot (0, y, z) = Ψ̃tot (W, y, z) .
 10

The transverse momentum k⊥ = (kx , ky ) is conserved, and its components are quantized. Imposing periodic boundary
conditions, we demand that kx = 2 πWnx ≡ qnx , ky = 2 πWnz ≡ qny , where n ≡ {nx , ny } ∈ Z. Along the z-axis, we seek
plane-wave solutions of the form ei kz z . Consequently, the full wavefunction takes the form of

Ψ̃tot (x, y, z, qnx , qny ) = const. × Ψ̃n (x) ei (qnx x+qny y) . (24)
q
For any mode of a given value of the transverse-momentum component, defined as k⊥ = qn2 x + qn2 y (with k⊥ ≤ E), we
can determine the z-component of the wavevectors of the incoming, reflected, and transmitted waves (denoted by kin ), by
2 2


solving 4 m2 E 2 = kz2 + k⊥ ⇒ kz2 = ± 2 m E − k⊥
2
. In fact, in the regions z < 0 and z > L, this relation leads to the
following four solutions:
q q
kz = ± 2 m E − k⊥ 2 and k = ± i 2 m E + k⊥2 . (25)
z

As we have understood by now from the 2d case, this implies the presence of the evanescent waves, corresponding to the
imaginary solutions for kz .
We will follow the same procedure as described for the 2d QBCP. Without any loss of generality, we consider the
transport of one of the degenerate positive energy states, Ψ+,1 , corresponding to electron-like particles for one of the two
degenrate conduction bands, with the Fermi level outside the potential barrier being adjusted to a value E = ε+ 3d (kx , ky , kz ).
i (qnx x+qny y)
In this case, a scattering state ∼ Ψ̃n (z) e , labelled by the vectorial index n, is constructed in a piecewise fashion
as

ϕ̃L (z)
 for z < 0
Ψ̃n (z) = ϕ̃M (z) for 0 ≤ z ≤ L , (26)

ϕ̃ (z)
R for z > L

where
Ψ+,1 (qnx , qny , kin ) ei kin z + rn,s Ψ+,s (qnx , qny , − kin ) e−i kin z
P
s=1,2 X
ϕ̃L (z) = q + r̃n,s Ψ−,s (qnx , qny , − i κ) eκ |z| ,
Ṽ(kin , n) s=1,2
"
X
ϕ̃M (z) = αn,s Ψ+,s (qnx , qny , kmid ) ei kmid z + βn,s Ψ+,s (qnx , qny , − kmid ) e−i kmid z
s=1,2
#
− κmid z κmid z
+ α̃n,s Ψ−,s (qnx , qny , i κmid ) e + β̃n,s Ψ−,s (qnx , qny , − i κmid ) e Θ (E − V0 )
"
X
+ αn,s Ψ−,s (qnx , qny , kmid ) ei kmid z + βn,s Ψ−,s (qnx , qny , − kmid ) e−i kmid z
s=1,2
#
− κmid z κmid z
+ α̃n,s Ψ+,s (qnx , qny , i κmid ) e + β̃n,s Ψ+,s (qnx , qny , − i κmid ) e Θ (V0 − E) ,
P
tn,s Ψ+,s (kin , qny , qnz )
s=1,2 X
ϕ̃R (z) = q ei kin (z−L) + r̃n,s Ψ−,s (qnx , qny , i κ) ei κ (z−L) ,
Ṽ(kin , n) s=1,2

kin
q q
kin = 2 m E − k⊥ 2 , Ṽ(kin , n) ≡ |∂kin ε3d (qnx , qny , kin )| = , κ = 2 m E + k⊥ 2 ,

q q m
kmid = 2 m |E − V0 | − k⊥ 2 , κmid = 2 m |E − V0 | + k⊥ 2 . (27)

Here, the magnitude of the group velocity, defined by Ṽ(kin , n), is needed to define the reflection and transmission
coefficients appearing the unitary scattering matrix.
The boundary conditions at z = 0 and z = L lead to

ϕ̃L (0) = ϕ̃M (0) and ∂z ϕ̃L (z) z=0

= ∂z ϕ̃M (x) z=0
;
ϕ̃M (L) = ϕ̃R (L) and ∂z ϕ̃M (z) z=L
= ∂z ϕ̃L (z) z=L
. (28)
 11

FIG. 7. 3d QBCP: The polar plots show the transmission coefficient, T (E, V0 , θ) E<V (upper panel), and the reflection coefficient
0
R(E, V0 , θ) E≤V (lower panel), as functions of the incident angle θ, for some representative values of 2 m L2 V0 . The values of E
0
are shown in the plot-legends.

These conditions are correspond to the continuity of the flux of the probability-current density along the z-direction.
From the matching of the wavefunction and its derivatives, we have 2 × 2 = 4 matrix-equations from the two bound-
aries. For 3d QPCBs, each of these matrix-equations can be separated into four components, since each wavevector
comprises four components. Therefore, we have 4 × 4 = 16 equations for the sixteen undetermined coefficients, captured
by {rn,1 , r̃n,1 , αn,1 , βn,1 , α̃n,1 , β̃n,1 , tn,1 , t̃n,1 } ∪ {rn,2 , r̃n,2 , αn,2 , βn,2 , α̃n,2 , β̃n,2 , tn,2 , t̃n,2 }. Again, since the explicit ana-
lytical expressions for tn,1 , tn,2 , rn,1 and rn,r are extremely long, we refrain from showing them here. Instead, we represent
their characteristics via Figs. 7 and 8.
The reflection and transmission coefficients at an energy E are given by

R(E, V0 , θ) = |rn,1 (E, V0 )|2 + |rn,2 (E, V0 )|2 and T (E, V0 , θ) = |tn,1 (E, V0 )|2 + |tn,2 (E, V0 )|2 , (29)

respectively, where θ = tan−1 (k⊥ /kin ) defines the incident angle of the incoming wave. Although we have not provided
the explicit expressions for tn,1 and tn,2 , rn,1 , one can check that |tn,2 (E, V0 )| = 0 when the incident state is assumed to
 12

FIG. 8. 3d QBCP: The polar plots show the transmission coefficient, T (E, V0 , θ) E>V (upper panel), and the reflection coefficient
0
T (E, V0 , θ) E>V (lower panel), as functions of the incident angle θ, for some representative values of 2 m L2 V0 . The values of E
0
are shown in the plot-legends.

be Ψ+,1 . 2 Figs. 7 and 8 illustrate the polar plots of T (E, V0 , θ) and R(E, V0 , θ), as functions of θ, for some representative
values of V0 , L, and E. If one wants to compare the results obtained for a 3d QBCP with those for a 3d electron-gas,
Eq. (14) and Figs. 4–5 are to be considered (i.e., the same as the 2d electron-gas).

B. Transmission coefficients, conductivity, and Fano factors

Again, we assume W to be very large such that qnx and qny can effectively be treated as continuous variables. Using
√ √ √
kin = 2 m E cos θ, kx = 2 m E sin θ cos ϕ, and ky = 2 m E sin θ sin ϕ, we get dkx dky = dϕ dθ m E | sin(2 θ)|. Hence, in

2 Similarly, |tn,1 (E, V0 )| = 0 when the incident state is assumed to be Ψ+,2 .

 13

(a)

(b)

FIG. 9. Plots of the conductivity (σ) and Fano factor (F ), as functions of E/V0 , for various values of 2 m L2 V0 . The subfigures (a)
and (b) correspond to 3d QBCP and 3d electron-gas, respectively. While σ has the units of eV, F is unitless.

the zero-temperature limit, the conductance is given by [cf. Eq. (1)]

Z √2 m E Z 2π
e2 X 2 e2 W 2
G(E, V0 ) = 2 × |tn,1 | → dk ⊥ k⊥ dϕ T (E, V0 , k⊥ )
2π n 2 π3 0 0

e2 2 W 2 (2 m E) π/2 2 e2 W 2 m E π/2
Z Z
= dθ | sin θ cos θ| T (E, V0 , θ) = dθ | sin(2θ)| T (E, V0 , θ) . (30)
π π 0 π2 0

In order to account for the twofold degeneracy (since we have two independent conduction bands), we have included an
extra factor of two. We note that the conductivity is given by

2 e2 L m E π/2
Z
L
σ(E, V0 ) = 2 G(E, V0 ) = dθ | sin(2θ)| T (E, V0 , θ) . (31)
W π2 0

In Fig. 9(a), we illustrate the conductivity and the Fano factor [using Eq. (2)], as functions of E/V0 , for some values
of 2 m L2 V0 . Side by side, the corresponding curves [generated using Eq. (14)] corresponding for a 2d electron-gas are
provided in Fig. 9(b) for the sake of comparison.

IV. Summary and outlook

In this paper, we have revisited the problem of the tunneling of quasiparticles for 2d and 3d QBCP semimetals, mainly
to correct the implementations of the boundary conditions by including parts featuring evanescent waves. The inclusion
of these evanescent waves, arising from a QBCP’s quadratic-in-momentum dispersion, were missed in our earlier work
 14

[11]. With the correct number of wavevector-solutions solutions in place, we have proceeded to compare the transport
characteristics (obtained using the Landauer-Büttiker formalism) with those of normal electron-gases. The results also
enable us to contrast with those for other nodal-point semimetals exemplified by graphene [32, 37], pseudospin-1 (or triple-
point) semimetals [38–40], Rarita-Schwinger-Weyl semimetals [40], and Weyl/multi-Weyl semimetals [25, 26, 41, 42]. The
conclusion that QBCPs do not exhibit Klein tunneling remains valid (where Klein tunneling is characterized by T = 1 at
normal incidence), unlike the Dirac quasiparticles of graphene, Weyl/multi-Weyl semimetals, and triple-point fermions.
Such differences can be observed by designing appropriate experimental set-ups involving the QBCPs. The extra evanescent
waves are the artifact of a dispersion whose momentum dependence along the propagation direction is of an order higher
than the linear order cases. Semi-Dirac semimetals [27, 28], bilayer graphene [32], and multi-Weyl semimetals [25, 26, 42]
provide examples of such nonlinear-in-momentum dispersive bands.
In the future, it will be worthwhile to determine the nature of Josephson junctions constructed out of the 2d and 3d
QBCPs, and determine the nature of the Andreev bound states with sub-gap energies. This exercise has been carried
out for graphene, Weyl/multi-Weyl semimetals, semi-Dirac nodal points, pseudospin-1 semimetals, and Rarita-Schwinger-
Weyl semimetals [28, 43–46]. The computation will be significantly harder than the nodal-point semimetals (considered
till now in the literature), because the presence of the extra evanescent waves (resulting from the quadratic-in-momentum
dispersion) will lead to solving polynomial equations resulting form the determinant of 8-dimensional and 16-dimensional
matrices for the 2d and 3d QBCPs, respectively, for a Josephson junction modelled as a delta-function potential [28].

[1] A. H. Castro Neto, F. Guinea, N. M. R. Peres, K. S. Novoselov, and A. K. Geim, The electronic properties of graphene, Rev.
Mod. Phys. 81, 109 (2009).
[2] N. P. Armitage, E. J. Mele, and A. Vishwanath, Weyl and dirac semimetals in three-dimensional solids, Rev. Mod. Phys. 90,
015001 (2018).
[3] K. Sun, H. Yao, E. Fradkin, and S. A. Kivelson, Topological insulators and nematic phases from spontaneous symmetry breaking
in 2D Fermi systems with a quadratic band crossing, Phys. Rev. Lett. 103, 046811 (2009).
[4] W.-F. Tsai, C. Fang, H. Yao, and J. Hu, Interaction-driven topological and nematic phases on the lieb lattice, New Journal of
Physics 17, 055016 (2015).
[5] I. Mandal and S. Gemsheim, Emergence of topological Mott insulators in proximity of quadratic band touching points, Con-
densed Matter Physics 22, 13701 (2019).
[6] A. A. Abrikosov, Calculation of critical indices for zero-gap semiconductors, JETP 39, 709 (1974).
[7] T. Kondo, M. Nakayama, R. Chen, J. J. Ishikawa, E. G. Moon, T. Yamamoto, Y. Ota, W. Malaeb, H. Kanai, Y. Nakashima,
Y. Ishida, R. Yoshida, H. Yamamoto, M. Matsunami, S. Kimura, N. Inami, K. Ono, H. Kumigashira, S. Nakatsuji, L. Balents,
and S. Shin, Quadratic fermi node in a 3d strongly correlated semimetal, Nat. Commun. 6 (2015).
[8] E.-G. Moon, C. Xu, Y. B. Kim, and L. Balents, Non-Fermi-liquid and topological states with strong spin-orbit coupling, Phys.
Rev. Lett. 111, 206401 (2013).
[9] R. M. Nandkishore and S. A. Parameswaran, Disorder-driven destruction of a non-Fermi liquid semimetal studied by renor-
malization group analysis, Phys. Rev. B 95, 205106 (2017).
[10] I. Mandal and R. M. Nandkishore, Interplay of Coulomb interactions and disorder in three-dimensional quadratic band crossings
without time-reversal symmetry and with unequal masses for conduction and valence bands, Phys. Rev. B 97, 125121 (2018).
[11] I. Mandal, Tunneling in Fermi systems with quadratic band crossing points, Annals of Physics 419, 168235 (2020).
[12] B. Cheng, T. Ohtsuki, D. Chaudhuri, S. Nakatsuji, M. Lippmaa, and N. Armitage, Dielectric anomalies and interactions in the
three-dimensional quadratic band touching Luttinger semimetal Pr2 Ir2 O7 , Nat. Commun. 8, 2097 (2017).
[13] I. Mandal and H. Freire, Transport in the non-Fermi liquid phase of isotropic Luttinger semimetals, Phys. Rev. B 103, 195116
(2021).
[14] H. Freire and I. Mandal, Thermoelectric and thermal properties of the weakly disordered non-Fermi liquid phase of Luttinger
semimetals, Physics Letters A 407, 127470 (2021).
[15] I. Mandal and H. Freire, Raman response and shear viscosity in the non-Fermi liquid phase of Luttinger semimetals, Journal
of Physics: Condensed Matter 34, 275604 (2022).
[16] I. Mandal, Search for plasmons in isotropic Luttinger semimetals, Annals of Physics 406, 173 (2019).
[17] J. Wang and I. Mandal, Anatomy of plasmons in generic Luttinger semimetals, Eur. Phys. J. B 96, 132 (2023).
[18] S. Bera and I. Mandal, Floquet scattering of quadratic band-touching semimetals through a time-periodic potential well, J.
Phys. Condens. Matter 33, 295502 (2021).
[19] D. Yanagishima and Y. Maeno, Metal-nonmetal changeover in pyrochlore iridates, Journal of the Physical Society of Japan 70,
2880 (2001).
[20] K. Matsuhira, M. Wakeshima, R. Nakanishi, T. Yamada, A. Nakamura, W. Kawano, S. Takagi, and Y. Hinatsu, Metal–insulator
transition in pyrochlore iridates Ln2 Ir2 O7 (Ln = Nd, Sm, and Eu), Journal of the Physical Society of Japan 76, 043706 (2007).
[21] A. A. Abrikosov and S. D. Beneslavskiǐ, Possible existence of substances intermediate between metals and dielectrics, Soviet
Journal of Experimental and Theoretical Physics 32, 699 (1971).
[22] J. M. Luttinger, Quantum theory of cyclotron resonance in semiconductors: General theory, Phys. Rev. 102, 1030 (1956).
[23] S. Murakami, N. Nagaosa, and S.-C. Zhang, SU(2) non-Abelian holonomy and dissipationless spin current in semiconductors,
Phys. Rev. B 69, 235206 (2004).
 15

[24] I. Boettcher and I. F. Herbut, Superconducting quantum criticality in three-dimensional Luttinger semimetals, Phys. Rev. B
93, 205138 (2016).
[25] Y.-H. Deng, H.-F. Lü, S.-S. Ke, Y. Guo, and H.-W. Zhang, Quantum tunneling through a rectangular barrier in multi-Weyl
semimetals, Phys. Rev. B 101, 085410 (2020).
[26] I. Mandal and A. Sen, Tunneling of multi-Weyl semimetals through a potential barrier under the influence of magnetic fields,
Physics Letters A 399, 127293 (2021).
[27] S. Banerjee, R. R. P. Singh, V. Pardo, and W. E. Pickett, Tight-binding modeling and low-energy behavior of the semi-Dirac
point, Phys. Rev. Lett. 103, 016402 (2009).
[28] I. Mandal, Andreev bound states in Josephson junctions of semi-Dirac semimetals, Physica B: Condensed Matter 683, 415918
(2024).
[29] R. Landauer, Spatial variation of currents and fields due to localized scatterers in metallic conduction (and comment), Journal
of Mathematical Physics 37, 5259 (1996).
[30] M. Büttiker, Quantized transmission of a saddle-point constriction, Phys. Rev. B 41, 7906 (1990).
[31] Y. Blanter and M. Büttiker, Shot noise in mesoscopic conductors, Physics Reports 336, 1 (2000).
[32] M. I. Katsnelson, K. S. Novoselov, and A. K. Geim, Chiral tunnelling and the Klein paradox in graphene, Nature Physics 2,
620 (2006).
[33] M. Salehi and S. Jafari, Quantum transport through 3D Dirac materials, Annals of Physics 359, 64 (2015).
[34] J. Tworzydlo, B. Trauzettel, M. Titov, A. Rycerz, and C. W. J. Beenakker, Sub-Poissonian shot noise in graphene, Phys. Rev.
Lett. 96, 246802 (2006).
[35] L. Janssen and I. F. Herbut, Nematic quantum criticality in three-dimensional Fermi system with quadratic band touching,
Phys. Rev. B 92, 045117 (2015).
[36] I. F. Herbut, Isospin of topological defects in dirac systems, Phys. Rev. B 85, 085304 (2012).
[37] P. E. Allain and J. N. Fuchs, Klein tunneling in graphene: optics with massless electrons, The European Physical Journal B
83, 301–317 (2011).
[38] A. Fang, Z. Q. Zhang, S. G. Louie, and C. T. Chan, Klein tunneling and supercollimation of pseudospin-1 electromagnetic
waves, Phys. Rev. B 93, 035422 (2016).
[39] R. Zhu and P. M. Hui, Shot noise and Fano factor in tunneling in three-band pseudospin-1 Dirac–Weyl systems, Physics Letters
A 381, 1971 (2017).
[40] I. Mandal, Transmission in pseudospin-1 and pseudospin-3/2 semimetals with linear dispersion through scalar and vector
potential barriers, Physics Letters A 384, 126666 (2020).
[41] C. Yesilyurt, S. G. Tan, G. Liang, and M. B. A. Jalil, Klein tunneling in Weyl semimetals under the influence of magnetic field,
Scientific Reports 6, 38862 (2016).
[42] I. Mandal, Transmission and conductance across junctions of isotropic and anisotropic three-dimensional semimetals, European
Physical Journal Plus 138, 1039 (2023).
[43] M. Maiti and K. Sengupta, Josephson effect in graphene superconductor/barrier/superconductor junctions: Oscillatory behavior
of the Josephson current, Phys. Rev. B 76, 054513 (2007).
[44] K. Kulikov, D. Sinha, Y. M. Shukrinov, and K. Sengupta, Josephson junctions of Weyl and multi-Weyl semimetals, Phys. Rev.
B 101, 075110 (2020).
[45] I. Mandal, Andreev bound states in superconductor-barrier-superconductor junctions of Rarita-Schwinger-Weyl semimetals,
Physics Letters A 503, 129410 (2024).
[46] I. Mandal, Nature of Andreev bound states in Josephson junctions of triple-point semimetals, arXiv e-prints (2024),
arXiv:2406.15350 [[Link]-con].