Signatures of topology in generic transport measurements for Rarita-Schwinger-Weyl

semimetals
Ipsita Mandal1 ,∗ Shreya Saha1 , and Rahul Ghosh2
1
Department of Physics, Shiv Nadar Institution of Eminence (SNIoE),
Gautam Buddha Nagar, Uttar Pradesh 201314, India
2
Department of Physical Sciences, Indian Institute of Science
Education and Research Berhampur, Berhampur, Odisha 760010, India

We investigate how the signatures of the topological properties of the bandstructures for nodal-
point semimetals are embedded in the response coefficients, arising in two distinct experimental
set-ups, by taking the Rarita-Schwinger-Weyl (RSW) semimetal as an example. The first sce-
nario involves the computation of third-rank tensors representing second-order response coefficients,
relating the charge/thermal current densities to the combined effects of the gradient of the chem-
ical potential and an external electric field/temperature gradient. On the premises that internode
scatterings can be ignored, the relaxation-time approximation leads to a quantized value for the non-
vanishing components of each of these nonlinear response tensors, characterizing a single untilted
RSW node. Furthermore, the final expressions turn out to be insensitive to the specific values of
the chemical potential and the temperature. The second scenario involves computing the magneto-
electric conductivity under the action of collinear electric (E) and magnetic (B) fields, representing
a planar Hall set-up. In particular, our focus is in bringing out the dependence of the linear-in-|B|
parts of the conductivity tensor on the intrinsic topological properties of the bandstructure, which
are nonvanishing only in the presence of a nonzero tilt in the energy spectrum.

I. INTRODUCTION

Three-dimensional (3d) topological semimetals [1–4] form the gapless cousins of the topological insulators (for which a
gap-opening implying a topological phase transition), possessing a gapless spectrum. Weyl semimetals (WSMs) exemplify
such topological phases, forming an intermediate state in the transition from metals to insulators, in which the conduction
and the valence bands touch only at discrete points. This leads to the emergence of zero band-gap and singular points in
the Brillouin zone (BZ), known as nodal points, where the density-of-states vanish exactly. The nodal are topologically
stable as they cannot be fully gapped out by perturbations that are small (in magnitude) and local in momentum space,
such that the bulk gap remains intact sufficiently away from the band-crossing points. The stability of the nodal points
and, consequently, the gaplessness of the semimetallic phases are ensured by the fact these points act as the sources/sinks of
topological charges (leading to the noting of defective or singular points). Although the topological charges are determined
by the electronic bandstructure of the material, often they are not directly measurable. However, two experimentally-
accessible and exploitable properties that appear are:
1. Quantized response in various transport measurements that depend on the intrinsic Berry curvature (BC) of the
bandstructures. Some examples include circular dichroism [5, 6], circular photogalvanic effect [7–10], and Magnus
Hall effect [11–13].
2. Fermi arcs surface states at the two-dimensional (2d) surface BZ [3].
Although not in the form of a quantized response, the signatures of BC are very much present in numerous other transport
properties, such as intrinsic anomalous Hall effect [14–16], planar Hall effects [17–36], magneto-optical conductivity with
strong (quantizing) magnetic field [37–39], transmission of quasiparticles across barriers/wells of electric potential [40–44].
In the WSMs [1, 2], we encounter twofold band-crossing singularities, with each band having an isotropic linear-in-
momentum dispersion in the vicinity of the nodal point. The 230 space groups in nonrelativistic condensed matter physics
allows the possibility of richer bandstructures in the form symmetry-protected multifold band-crossings [45]. The simplest
case is the one where each band exhibits an isotropic linear-in-momentum dispersion, akin to the WSMs. In general, the
effective low-energy Hamiltonian of such a system is captured by ∼ k · S, where S is the vector operator comprising the
three components of the angular momentum in the spin-ς representation of the SO(3) group. This stems from the (2 ς + 1)
bands touching at the nodal point, resulting in the itinerant electronic degrees of freedom represented by quasiparticles
with pseudospin quantum numbers equalling ς. The terminology of “pseudospin” is used in order to clearly demarcate it
as a quantum number distinct from the relativistic spin of an electron. While WSMs feature ς = 1/2, the poster child
for multifold cases is the Rarita-Schwinger-Weyl (RSW) semimetal [5, 13, 36, 44–53], having pseudospin-3/2 (i.e., fourfold
band-crossings). This nomenclature is inspired by the fact that in the branch of theoretical high-energy physics, the

∗ [email protected]
2

(a) (b)

FIG. 1. Schematics of the dispersions of the four bands at a single RSW node: Subfigures (a) and (b) represent an untilted node
(i.e., η = 0) and a tilted node (with 0 < |η| < 1), respectively.

Rarita-Schwinger (RS) equation describes the field equation of elementary particles with the relativistic spin of 3/2, which
naturally arise in supergravity models [54]. Albeit, these higher-spin fermionic particles appear neither in the standard
model of particle physics, nor has been detected experimentally. The RSW semimetals thus present a nonrelativistic
analogoue of the elusive RS fermions in the context of solid-state systems.
Let us now elaborate on the origins of the BC and the quantized-response phenonmena in widely different experimental
settings. The topological properties of a crystalline bandstructure are inferred by treating the BZ as a closed manifold.
A nodal-point semimetal is endowed with a nontrivial topology when the nodes demarcate point-like topological defects,
mathematically quantified as the locations of the BC monopoles [4, 55]. The charge of a BC monopole, therefore, is a
topological charge, whose sign assigns a chirality χ to the node, with χ = +1 and χ = −1 labelling the so-called right-
moving and left-moving chiral quasiparticles, respectively. Here, we will adopt the widely-used convention of assigning
χ = 1 to the bands with negative energy (with respect to the band-crossing point). The physical picture thus represents
that a positively-charged (negatively-charged) monopole acts as a source (sink) for the BC flux lines. Using all the above
definitions, we find that the monopole charge of a specific band is computed by employing the familiar Gauss’s law, where
we integrate the BC flux over a closed 2d surface enclosing the point-defect. If we project on to the space of a pair of bands
with the same magnitide of the pseudospin magnetic quantum numbers, we get a two-level system — the Chern number
(C) represents a wrapping number of the map from the 2d closed surface (topologically equivalent to S 2 ) to the Bloch
sphere (S 2 ), given by the elements of the second homotopy group Π2 (S 2 ) = Z. Thus the monopole charges are equivalent
to the Chern numbers, when interpreted as the wrapping numbers of the defects. The monopole charges points always
appear in pairs, with each pair having the values ± C, thus satisfying the Nielson-Ninomiya theorem [56]. Our aim in
this paper is to unravel some transport coefficients for an RSW node which show a quantized nature, being proportional
to C, thus revealing the nature of the underlying topology. In this context, it is necessary to point out that the four
bands at a single RSW node have Chern numbers ±1 and ±3, which sum up to a net monopole charge of magnitude 4
(considering either the two upper or the two lower bands). Therefore, while a WSM-node harbours a Berry monopole
charge of magnitude unity, an RSW hosts a net monopole charge of a higher-integer value. In this paper, we will show how
these higher values of charges may show up as quantized response in various experimental set-ups. In fact, the signatures
of the existence of RSW quasiparticles are reflected by the large values of the topological charges existing in a range of
materials, such as CoSi [50, 57], RhSi [58], AlPt [59], and PdBiSe [60].
In this paper, we consider the quantization associated with two distinct experimental set-ups:
1. Nonlinear transport under the effect of an external electric field (and/or temperature gradient) and the gradient of
the chemical potential, constituting a response of electrochemical (or thermochemical) nature [61]. Here, we dub the
associated electrical and thermal conductivity tensors as the electrochemical response (ECR) and thermochemical
response (TCR), respectively. For a single untilted node of WSM, the ECR has been shown to take quantized values
for temperature T equalling zero [61]. However, there were various shortcomings in their steps to derive the form of
the response, which we will clarify in the course of our computations. Moreover, we will show the quantized nature
of both the ECR and the TCR considering a generic temperature 1 for the fourfold nodal point of the RSW, which
is a multiband generalization of the WSM case.

1 Albeit with the constraint T ≪ µ (while using natural units), so that we are in the low-energy limit where the effective continuum model for
the node is valid. Since µ ranges around 0.01-0.1 eV [50], we must have T ≲ 10−3 eV. For example, recent experiments [62] probe response
setting the temperature ranges of the order of 8.617 × 10−5 – 8.617 × 10−3 eV, which is equivalent to ∼ 1 – 100 K (in SI units).
3

2. Components of the magnetoelectric conductivity under the action of electric (E) and magnetic (B) fields, applied
parallel to each other, constituting a planar Hall set-up. Here, we will demonstrate the quantized nature of the
linear-in-B parts of the linear-response tensor, arising from the four bands of an RSW node.

The paper is organized as follows. In Sec. II, we outline the form of the low-energy effective Hamiltonian in the vicinity
of an RSW node and, also, show the expressions for various topological quantities. While Sec. III deals with the nonlinear
response coefficients dubbed as ECR and TCR, Sec. IV focusses on the linear response associated with the magnetoelectric
conductivity. In the end, we wrap up with a summary and outlook in Sec. V. In Appendix A, the details of the derivations
of the response tensors, applying the semiclassical Boltzmann formalism, are provided. The remaining appendices are
devoted to elaborating on some of the details of the intermediate steps, necessary to derive the final expressions shown
in the main text. In all our expressions, we resort to using the natural units, which implies that the reduced Planck’s
constant (ℏ), the speed of light (c), and the Boltzmann constant (kB ) are each set to unity.

II. MODEL

With the help of group-theoretic symmetry analysis and first principles calculations, it has been shown that seven space
groups may host fourfold band-crossing points [45] at high-symmetry points of the BZ. Nearly 40 candidate materials have
also been identified that can host the resulting RSW quasiparticles. Ref. [63] has tabulated the multifold degneracies in the
65 chiral space groups characterizing the chiral crysals, which are the ones with only orientation-preserving symmetries. A
chiral fourfold band-crossing can be realized in the space groups (1) 195–198 and 207–214 at the Γ-point; (2) 207 and 208
at the R-point; and (3) 211 and 214 at the H-point, in the presence of spin-orbit coupling. These fourfold degeneracies
exhibit a BC texture that is homotopic to that of a spin-3/2 moment in a magnetic field. The RSW semimetal, with the
effective Hamiltonian possessing a full SU(2) invariance (i.e., a full rotational invariance). It has been shown [64] that
chiral topological metals belonging to the SrGePt family (e.g., SrSiPd, BaSiPd, CaSiPt, SrSiPt, BaSiPt, and BaGePt),
characterized by the space group 198, host RSW quasiparticles, sixfold excitations (two copies of pseudospin-1 fermions), as
well as Weyl points in their bandstructures, when spin-orbit coupling is considered. More explicitly, a fourfold-degenerate
node appears at the center of the BZ (i.e., the Γ-point), carrying the monopole charge of + 4, while a sixfold-degenerate
node arises at the boundary of the BZ (i.e., the R-point) with a net monopole charge equalling − 2 − 2 = − 4. Here, we
ignore any internode scatterings, which is justified because the energy offset between the fourfold-degenerate point (at Γ)
and the sixfold-degenerate point (at R) is usually large [50, 64–66].
The usual method of linearizing the k · p Hamiltonian about such a degeneracy point provides us with the low-energy
effective continuum Hamiltonian, valid in the vicinity of the node. The explicit form of this Hamiltonian, for a single node
with chirality χ, is given by

H(k) = η v0 kz + χ v0 (kx Jx + ky Jy + kz Jz ) , (1)

where J = {Jx , Jy , Jz } represents the vector operator whose three components comprise the angular momentum oper-
ators in the spin-3/2 representation of the SO(3) group. Here, η represents the tilt parameter, with |η| < 1 representing
the type-I phase, which is the scenario under consideration. We choose the commonly-used representation where
 √   √ 
3 −i 3
0 0 0 0 0 0
3 
√ 2 √ 2 2 00 0
 3 i 3 0 1 0
0 1 0 0 −i 0√  0 
 
Jx =  2 √ , Jy =  2 , Jz =  2
0 0 − 1 0  . (2)
  
0 3 −i 3 
1 0
√ 2
 0 i 0
√ 2
 2
0 0 3
0 0 0 i 3
0 0 0 0 − 23
2 2

Our convention is such that the pair of conjugate nodes are separated along the kz -direction. The energy eigenvalues are
found to be

1 3
εs (k) = s v0 k + v0 η kz , s∈ ± ,± , (3)
2 2
q
where k = kx2 + ky2 + kz2 . Hence, each of four bands has a linear-in-momentum dispersion (cf. Fig. 1). The signs of “+”
and “−” give us the dispersion relations for the conduction and valence bands, respectively. The corresponding group
velocities of the quasiparticles are given by

s v0 k
vs (k) = ∇k εs (k) = + η v0 ẑ . (4)
k
4

A. Properties with topological origins

A nontrivial topology of the bandstructure of the RSW semimetals gives rise to the BC and the orbital magnetic moment
(OMM), using the starting expressions of [67]
e
Ωs (k) = i ⟨∇k ψs (k)| × |∇k ψs (k)⟩ and ms (k) = − Im [⟨∇k ψs | × (H(k) − εs (k)) |∇k ψs ⟩] , (5)
2
respectively. Here, |ψs (k)⟩ denotes the eigenfunction for the sth band at the node with chirality χ, and e denotes the
magnitude of the charge of a single electron. Evaluating these expressions for the RSW node described by H(k), we get
χsk e χ v0 Gs k 3 7
Ωs (k) = − 3
and ms (k) = − 2
, with G± 23 = and G± 12 = . (6)
k k 4 4
Since Ωs (k) and ms (k) are the intrinsic properties of the bandstructure, they depend only on the wavefunctions. Clearly,
they are related as
e v0 Gs k
ms (k) = Ωs (k) . (7)
s
We observe that, unlike the BC, the OMM does not depend on the sign of the energy dispersion. Since the tilt parameter
enters the Hamiltonian as the coefficient of the 4 × 4 identity matrix, its absence does not affect the eigenspinors and,
hence, the topology of the low-energy theory in the vicinity of the band-crossing point. Clearly, the BC or OMM of an
RSW node does not depend on the tilt parameter.
The OMM behaves exactly like the electron spin, because, on applying a magnetic field B, it couples to the field through
a Zeeman-like term, quantified by
k·B
ζs (k) = − ms (k) · B = e χ v0 Gs . (8)
k2
Therefore, we have

B − 2 k̂ k̂ · B
ξs (k) = εs (k) + ζs (k) , ws (k) = vs (k) + us (k) , us (k) = ∇k ζs (k) = e χ v0 Gs , (9)
k2
where ξs (k) and ws (k) are the OMM-modified energy and band velocity of the quasiparticles, respectively. With the
usual usage of notations, k̂ is the unit vector along k. The full rotational isotropy of the Fermi surface, for each band of
the RSW node, is broken by the inclusion of the OMM corrections.

B. Chern numbers

Using Eq. (6), the Chern number for the sth band can be evaluated as
Z
1
Cs = dS · Ωs , (10)
2π S
where S denotes a closed surface enclosing the k = 0 point and dS denotes the outwardly-directed area vector for an
infinitesimal patch on S. Exploiting the spherical symmetry of the Fermi surfaces for the bands of an isotropic RSW
node, we can choose S to be the surface of a unit sphere. Expressing in terms of the spherical polar coordinates, we use
kx = sin θ cos ϕ, ky = sin θ sin ϕ, and kz = cos θ, which gives us dS = sin θ dθ dϕ k̂. This leads to
Z
χs
Cs = − dθ dϕ sin θ = − 2 χ s . (11)
2π S

III. QUANTIZATION OBSERVED IN NONLINEAR RESPONSE

In this section, we will unravel the quantized nature of the ECR and the TCR, associated with the electric and thermal
currents, respectively, as explained in the introduction. The reader is referred to Appendix A for a detailed derivation of
the forms of the electric and thermal current densities, denoted by Js and Jth s , respectively, starting from the Boltzmann
equations. The final expressions are obtained by solving for δfs (r, k) [defined in Eq. (A15)], which is the deviation of the
distribution of the quasiparticles from the equilibrium, upto second order in ϵ. Here, ϵ ∈ [0, 1] is perturbative parameter
which quantifies the smallness of the magnitude of the probe fields comprising E, ∇r T , and ∇r µ. In other words, we work
in the regime where |E| ∝ ϵ, |∇r T | ∝ ϵ, and |∇r µ| ∝ ϵ, and use the expression
fs (r, k) = f0 (εs ) + ϵ fs(1) (εs ) + ϵ2 fs(2) (εs ) + O(ϵ3 ) (12)
in a perturbative expansion. Finally, we set ϵ = 1 at the end of our computations.
5

A. Electrochemical response

For computing the ECR, we set B and ∇r T to zero. The part of the total electrical current density (Js ), which shows a
quadratic dependence on the probe fields E and ∇r µ, is denoted by J̄s . This is the electrochemical current density, with
its components being proportional to the quadratic combinations of the form Ea ∂b µ. Its ath components is given by

J¯as = ϑsabc ∂b µ Ec , (13)

where ϑsabc is a rank-three conductivity tensor for the sth band (associated with the node of chirality χ). The dependence
of the electric current on the the probe fields at second order is the reason why we call the response to be nonlinear.
For computing ϑsabc , we find that the solutions turn out to be

fs(1) (εs ) = e τ (vs · E) f0′ (εs ) , fs(2) (εs ) = e τ [(E × Ωs ) · ∇r µ] f0′ (εs ) , (14)

where
∇r µ(r)
E(r) = E − . (15)
e
Due to the missing of the term ∇r µ in the Boltzmann equations by the authors in Ref. [61] [see Eq. (A5) for more details],
our solutions differ from theirs. Plugging in the solution in Eq. (A11), we obtain

d3 k
Z
s
J̄ = − e τ 2
[(e E − ∇r µ) × Ωs (vs · E) + vs {(E × Ωs ) · ∇r µ}] f0′ (εs ) (16)
(2 π)3

as the final expression for the response which is second-order in the probe fields. We divide it up as

J̄s = J(s,1) + J(s,2) + J(s,3) , (17)

where
d3 k d3 k
Z Z
J (s,1)
=e τ 2
(∇r µ × Ωs ) (vs · E) f0′ (εs ) , J(s,2) = − e2 τ vs [(E × Ωs ) · ∇r µ] f0′ (εs ) ,
(2 π)3 (2 π)3
e2 τ Gs d3 k
Z
J(s,3) = εs (vs · E) (Ωs × ∇r µ) f0′′ (εs ) . (18)
s2 (2 π)3
(3)
Here, Js represents the part arising from the magnetization density. We include it here, although it does not contribute
to transport measurements, to outline its quantized nature (in the absence of tilt). On analyzing the form of J(s,2) , we
find that a nonzero response is obtained when E is oriented perpendicular to ∇r µ, and only the component of Ωs , which
is perpendicular to both E and ∇r µ, will contribute. Now, for an untilted RSW node, we find that vs ∝ Ωs ∝ k̂. This
immediately tells us that the integral will give a nonzero answer only from the component of Ωs which is parallel to
(s,2) (s,1)
∝ ϵabc d3 k (vs )a (Ωs )a ∂b µ Ec . An analogous argument for J(s,1) tells us that Ja
R
the Rcurrrent density, i.e., Ja ∝
3
ϵabc d k (vs )a (Ωs )a ∂b µ Ec .
The corresponding components of the third-rank conductivity tensors are given by

d3 k d3 k
Z Z
(s,1) (s,2)
X X
′
ϑabc = e2 τ ϵabd3
(v )
s c (Ω ) f
s d 0 s(ε ) , ϑabc = − e 2
τ ϵ bcd (vs )a (Ωs )d f0′ (εs ) ,
(2 π) (2 π)3
d d
2 3
e τ Gs
Z
(s,3)
X d k (s,1) (s,2) (s,3)
and ϑabc = ϵadb εs (vs )c (Ωs )d f0′′ (εs ) , such that ϑsabc = ϑabc + ϑabc + ϑabc . (19)
s2 (2 π)3
d

For an untilted RSW node, we find that vs ∝ Ωs ∝ k̂. This immediately tells us that the integral must be proportional
(s,1) (s,3) (s,2)
to (1) δcd for ϑabc and ϑabc ; (2) δad for ϑabc . This gets rid of the summation over d, leading to

(s,1) (s,2) (s,3) e2 τ Gs

ϑabc = e2 τ ϵabc Is,c
(1)
, ϑabc = − e2 τ ϵabc Is,a
(1)
, ϑabc = − (2)
ϵabc Is,c , (20)
s2
where
d3 k d3 k
Z Z
(1)
Is,a ≡ (vs )a (Ωs )a f0′ (εs ) and Is,a
(2)
≡ εs (vs )a (Ωs )a f0′′ (εs ) . (21)
(2 π)3 (2 π)3
6

We employ the coordinate transformations shown in Eq. (B2). Using Eqs. (11), (B5), (B6), and (B7), we conclude that
(1) (1) (1) (1) (2) (2) (2) (2)
Is,x = Is,y = Is,z = Is /3 and Is,x = Is,y = Is,z = Is /3 where

Cs
Is(1) = − and Is(2) = − Is(1) . (22)
(2 π)2

Plugging these in, we find the final expressions as follows:

(s,1) (s,2) ϵabc e2 τ (1) (s,3) e 2 τ Gs

ϑabc = − ϑabc = Is , ϑabc = ϵabc Is(1) . (23)
3 3 s2
Therefore, each of the response coefficients is proportional to the Chern number, irrespective of the values of µ and T , and,
hence, show a completely quantized nature. Also, the nonzero response coefficients are the ones which are proportional
to ϵabc , representing the condition that J̄s ∝ E × ∇r µ (i.e., the ones arising when the components of the electric current
vector, E, and ∇r µ are oriented orthogonal to each other). It is important to note that for nonzero tilt, although the
leading-order terms (on expanding in |η| ≪ 1) take the quantized forms as shown above, nonzero subleading terms are
generated, which are η-dependent. A nonzero η also generates nonzero components beyond those ∝ ϵabc .

B. Thermochemical response

For computing the TCR, we set B and E to zero. The part of the total thermal current density (Jth s ), which shows a
quadratic dependence on the probe fields ∇r T and ∇r µ, is denoted by J̄th,s . This is the thermochemical current density,
with its components being proportional to the quadratic combinations of the form − ∂a T ∂b µ. Its ath components is given
by

J¯ath,s = φsabc ∂b µ (− ∂c T ) , (24)

where φsabc is a rank-three thermal-conductivity tensor for the sth band (associated with the node of chirality χ). The
dependence of the thermal current on the the probe fields at second order is the reason why this response is nonlinear.
For computing δϕsabc , we find that the solutions turn out to be

τ (εs − µ) τ (εs − µ)
fs(1) (εs ) = (vs · ∇r T ) f0′ (εs ) , fs(2) (εs ) = [(∇r T × Ωs ) · ∇r µ] f0′ (εs ) . (25)
T T
Analogous to the ECR case, we get

J̄th,s = Jth,(s,1) + Jth,(s,2) + Jth,(s,3) , (26)

where

d3 k
Z
τ 2
Jth,(s,1) = − (εs − µ) (∇r µ × Ωs ) (vs · ∇r T ) f0′ (εs ) ,
T (2 π)3
d3 k
Z
τ 2
Jth,(s,2) = (εs − µ) [vs (∇r T × Ωs ) · ∇r µ] f0′ (εs ) ,
T (2 π)3
τ Gs d3 k
Z
Jth,(s,3) =− 2 (εs − µ) εs (vs · ∇r T ) (Ωs × ∇r µ) [(εs − µ) f0′′ (εs ) + f0′ (εs )] . (27)
s T (2 π)3

Here, Jth,(s,3) represents the part arising from the magnetization density.
The corresponding components of the third-rank thermal conductivity tensors are given by

d3 k
Z
(s,1) τ X 2
φabc = (εs − µ) ϵabd (vs )c (Ωs )d f0′ (εs ) ,
T (2 π)3
d
d3 k
Z
(s,2) τ X 2
φabc =− (εs − µ) ϵbcd (vs )a (Ωs )d f0′ (εs ) , and
T (2 π)3
d
τ Gs X d3 k
Z
(s,3)
φabc = 2
(εs − µ) εs ϵadb (vs )c (Ωs )d [(εs − µ) f0′′ (εs ) + f0′ (εs )] ,
Ts (2π)3
d
(s,1) (s,2) (s,3)
such that φsabc = φabc + φabc + φabc . (28)
7

(s,1)
The relations vs ∝ Ωs ∝ k̂ for an untilted node again constrain the integrals to be proportional to (1) δcd for φabc and
(s,3) (s,2)
φabc ; (2) δad for φabc . Hence, we get rid of the summation over d, leading to

(s,1) τ ϵabc (3) (s,2) τ ϵabc (3) (s,3) τ Gs ϵabc (4)

φabc = Is,c , φabc = − Is,a , φabc = − Is,c . (29)
T T T s2
where
d3 k d3 k
Z Z h i
2 2 ′′
(3)
Is,a ≡ (εs − µ) (vs )a (Ωs )a f0′ (εs ) and Is,a
(4)
≡ ε s (v s )a (Ω s )a (ε s − µ) f 0 (ε s ) + (εs − µ) f ′
0 (ε s ) . (30)
(2 π)3 (2π)3
(3) (3) (3) (3) (4) (4) (4)
Employing Eqs. (B2), (11), (B5), (B6), and (B7), we conclude that Is,x = Is,y = Is,z = Is /3 and Is,x = Is,y = Is,z =
(4)
Is /3 where

π 2 T 2 Cs π 2 T 2 (1) 2 π 2 T 2 (1)
Is(3) = − = I s and I (4)
s = − Is . (31)
3 × (2 π)2 3 3

Plugging these in, we find the final expressions as follows:

(s,1) (s,2) π2 T (s,3) 2 π 2 T τ Gs

φabc = −φabc = ϵabc τ Is(1) , φabc = ϵabc Is(1) . (32)
3 3 s2
Clearly, the results are quantized only at a fixed value of T , as the overall response increases linearly with increasing T .
Needless to say here that the nonzero response coefficients are the ones which are proportional to ϵabc , representing the
condition that J̄th,s ∝ ∇r T × ∇r µ (i.e., the ones arising when the components of the thermal current vector, ∇r T , and
∇r µ are oriented orthogonal to each other). Similar to the ECR case, for a nonzero tilt, in addition to the leading-order
terms (on expanding in |η| ≪ 1) taking the quantized forms as shown above, η-dependent terms are generated. A nonzero
η will also be responsible for giving rise to nonzero components beyond those ∝ ϵabc .
From Eqs. (23) and (32), we find that

(s,i) 3 e2 (s,i)
ϑabc = φ (33)
π 2 T abc
(s,3) 2 (s,3)
is satisfied for i = 1 and i = 3, which embodies the Wiedemann-Franz law [68]. However, ϑabc = 23πe2 T φabc . This is
fine because the Wiedemann-Franz law is primarily applicable for linear response, and it is not universally valid in the
nonlinear regimes (see, for example [69]). A few more comments are in order: Let us define the scattering probability
(Wk,q ) as the probability that a quasiparticle with momentum k is scattered into any of the quantum states contained in
the infinitesimal volume d3 q, centred around q (assuming those states are unoccupied in order to satisfy Pauli principle).
At sufficiently low temperatures, scatterings caused by impurities are the dominant source in any realistic material [68]. If
the impurity-concentration is sufficiently low and the electron-impurity coupling is sufficiently weak, the actual derivation
of the collision integral shows that the scattering is necessarily inelastic. This leads to the scattering probabality Wk,q to
be essentially zero unless εs (k) = εs (q). Under these circumstances, the Wiedemann-Franz law holds [68]. However, for
inelastic scattering processes, when we can have a nonvanishing Wk,q for εs (k) ̸= εs (q), the Wiedemann-Franz law is not
expected to hold generically. However, such detailed analysis can be performed only by going beyond the relaxation-time
approximation, which is beyond the scope of work — hence, we leave it for future investigations.

IV. MAGNETOELECTRIC CONDUCTIVITY

In a planar Hall set-up comprising static uniform electric and magnetic fields applied parallel to each other, a nonzero
tilt induces the linear response from an RSW node to contain parts which vary linearly with B (which we demonstrate
below). In this section, our aim is to elucidate how the characteristic topological quantities arise in those linear-in-B parts
of the magnetoconductivity tensor.
In a planar Hall set-up, the magnetoelectric conductivity tensor is obtained from Eq. (A11), with the solution (see
Ref. [36] for detailed derivations starting from the Boltzmann equations)
∞
gs (k) X n
fs (k) = f0 (ξs ) + δfs (k) , δfs (k) = [−f0′ (ξs )] gs (k) , =− (e τ Ds ) L̂n [Dsχ (wsχ + Wsχ ) · E] , (34)
eτ n=0

where

L̂ = (wsχ × B) · ∇k (35)
8

η

2 2 −1
P − η Cs P χ Gs η s−(s −η ) tanh s
FIG. 2. Plots for F1 ≡ 4
(blue curve), F2 ≡ 2 η2
(orange curve), and Ftot ≡ F1 −
s=1/2,3/2 s=1/2,3/2
ςas .
P
4 F1 + F2 (dashed cyan curve) as functions of η, after setting χ = +1. We note that Ftot corresponds to the the sum
s=1/2,3/2
The behaviour for χ = −1 is obtained simply as the negative of the above curves.

is referred to as the Lorentz operator (because it includes the classical effect due to the Lorentz force). We would like to
point out here that, in Ref. [36], two of us have studied the question of linear response for planar Hall and planar thermal
set-ups consisting of untilted RSW nodes (i.e., η = 0). As such, the response coefficients did not have any part having
a linear-in-B dependence. The fact that the linear-in-B parts go to zero for η = 0 [36] is expected because, according
to the Onsager-Casimir reciprocity relations [33, 70–72], terms containing odd powers of B cannot arise in this situation.
However, in the presence of a nonzero tilt, the conditions are relaxed, making it possible for the existence of terms varying
linearly with B [33].
Here, we will consider that case when E and B are parallel to each other (see Ref. [73] for the same set-up applied to
WSMs). Therefore, if E is applied along the unit vector êE , such that E = E êE , we have B = B êE and ζs = − (ms )b B =
e χ v0 Gs kkb 2B [cf. Eq. (8)].
We are primarily interested in finding the form of the part of the magnetoelectric conductivity

tensor, denoted by σslin ab , which is linear-in-B. Expressing the corresponding part of the current as Jalin,s = σslin ab E b ,
let us define a third-rank tensor

lin c
∂ σslin ab
ςs ab
= . (36)
∂Bc
For our scenario with E ∥ B, we will be dealing with the vector defined as
X c X b
ςas = ςslin ab δ b c = ςslin ab ≡ ςa(s,1) + ςa(s,2) + ςa(s,3) + ςa(s,4) + ςa(s,5) , (37)
b,c b

where (see Appendix C for the intermediate steps)

d3 k δza η v0 e3 τ Cs
Z
ςa(s,1) = e3 τ 3
(vs )a (vs · Ωs ) f0′ (εs ) = − , ςa(s,2) = − 4 ςa(s,1) ,
(2 π) 4 π2
−1 η

e3 τ v0 Gs d3 k δza χ v0 e3 τ Gs s2 − η 2 tanh s −ηs
Z
(s,3) ′
ςa = (vs )a [∇k · (k Ωs )] f0 (εs ) = − for |η| < |s| ,
s (2 π)3 2 π2 η2
e3 τ v0 Gs d3 k e3 τ v0 Gs d3 k
Z Z
′
ςa(s,4) = [v s · ∂ka (k Ω s )] f 0 (ε s ) = ς (s,3)
a , ςa
(s,5)
= (vs )a k (vs · Ωs ) f0′′ (εs ) = − ςa(s,3) .
s (2 π)3 s (2 π)3
(38)
(s,1)
Appendix C 4 provides more details regarding the steps leading to the final expressions shown above. Here, ςa and
(s,2) (s,3) (s,4) (s,5)
ςa represent the parts arising purely from the BC (i.e., with no contribution from the OMM), ςa , ςa , and ςa
are the terms which go to zero if OMM is neglected. Some comments are in order:
1. We find that all the above response coefficients go to zero for η = 0, validating our arguments that the linear-in-B
terms do not survive in the limit of zero tilt.
(s,1) (s,2)
2. Since ςa and ςa are proportional to η Cs , not depending on µ and T , they reflect a robust quantized response.
Hence, it is easy to identify this part of the response irrespective of the specific material one chooses to study in
experiments. The corresponding linear-in-η behaviour is illustrated in Fig. 2 (blue line).
9

(s,3) (s,3) (s,4)

3. Although ςa , ςa , and ςa are proportional to χ Gs , and are independent of µ and T , they have a complicated
(nonlinear) dependence on η. Since η a nonuniversal parameter, changing its value according to the material under
consideration, it is then hard to disentangle/extract the feature of Gs -proportionality. The corresponding nonlinear
behaviour is illustrated in Fig. 2 (orange curve).
Despite the fact that it is difficult to disentangle and decipher the presence of Cs and Gs in the magnetoconductivity
in real experiments, our motivation here is strictly to understand theoretically how these quantities determine the net
response. In other words, we believe that our theoretical results illustrate how topological semimetals have the signatures
of topology robustly embedded in the response functions.

V. SUMMARY AND FUTURE PERSPECTIVES

In this paper, we have considered the response coefficients in two distinct experimental set-ups, with the intent of
identifying clear signatures of the topological features of the bandstructures of nodal-point semimetals, by taking RSW
semimetal as an example. In the first case, the response is nonlinear in nature, arising under the combined effects of
an external electric field (and/or temperature gradient) and the gradient of the chemical potential. When internode
scatterings can be ignored, the relaxation-time approximation leads to the ECR and TCR taking a quantized value for a
single untilted RSW node, insensitive to the specific values of the chemical potential and the temperature. The second case
investigates the linear-response coefficient in the form of the magnetoelectric conductivity under the action of collinear
electric and magnetic fields, resulting in a planar Hall set-up. In particular, we have considered the nature of the linear-
in-B parts of the conductivity tensor, which are nonvanishing only in the presence of a nonzero tilt. We find that the
components show a quantized nature with coefficients which are functions of η. In our calculations, we have assumed the
same relaxation times to be applicable for all the bands.
The advantage of considering RSW semimetals is that it provides a richer structure for demonstrating the topological
features contained in the response coefficients, compared to the WSMs, because of the fact that the former consists
of four bands (rather than just two). Basically, each band has its distinct topological features, reflected by the band-
dependent values of the BC and the OMM. Through our band-dependent results for the relevant response coefficients,
we have, thus, succeeded in chalking out the signatures of these individual bands. In the future, it will be worthwhile to
calculate the behaviour of the nonlinear response coefficients in situations where internode scatterings are included too
[74]. Furthermore, we would like to improve our calculations by going beyond the relaxation-time approximation, which
involves actually computing the collision integrals for all relevant scattering processes [30] (involving both interband and
interband interactions), rather than just using phenomenological values of momentum-independent relaxation times. This
will help us to make quatitative predictions for realistic scenarios, because disorder-induced scatterings will give rise to
all possible generic processes. Last but not the least, it will be interesting to investigate the effects of anisotropy/disorder
[5, 6, 75] and/or strong interactions on the quantized nature of the response [10].

DATA AVAILABILITY STATEMENT

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

Appendix A: Response resulting from the Boltzmann equations

In this appendix, we will outline the semiclassical Boltzmann equation (BE) formalism [32, 34, 36], which is used
here for determining the transport coefficients. This framework is applicable in the regime of small deviations from the
equilibrium quasiparticle distribution. If there is an externally-applied magnetic field B, then we will we assume that it
is of small in magnitude, associated with a small cyclotron frequency ωc = e B/(m∗ c) (where m∗ is the effective mass
with the magnitude ∼ 0.11 me [76], with me denoting the electron mass). This condition is necessary when we want to
focus on the regime where quantized Landau levels need not be considered, given by the inequality ℏ ωc ≪ µ, where µ is
the chemical potential cutting the energy band(s), thus defining the Fermi-energy level. Here, we will approximate the
form of the collision integral with a relaxation time (τ ), which is a momentum-independent phenomenological parameter.
Furthermore, we will focus on the limit when only intranode and intraband scatterings dominate in the collision processes,
such that τ corresponds exclusively to the intranode scattering time.2 Under these approximations, it is sufficient to derive
the relevant expressions for a single node, whose chirality is denoted by χ. The neglect of interband scatterings is justified
if only pseudospin-conserving processes are allowed. We note that the neglect of interband scatterings is justified if only
those scattering processes are allowed which preserve pseudospin-conserving processes.

2 We would like to mention here that the effects of internode scatterings in the linear response for multifold semimetals have been addressed
in Ref. [74], using again the simple relaxation-time approximation.
10

For a 3d system, we define the distribution function fs (r, k, t) for the quasiparticles occupying a Bloch band labelled by
3
the index s, with the crystal momentum k and dispersion εs (k), such that dNs = gs fs (r, k, t) (2d π)k3 d3 r is the number of
3
particles occupying an infinitesimal phase space volume of dVp = (2d π)k3 d3 r, centered at {r, k} at time t. Here, gs denotes
the degeneracy of the band. We define the OMM-corrected dispersion and the corresponding modified Bloch velocity as

ξs (k) = εs (k) + ζs (k) and ws (k) = ∇k εs (k) + ∇k ζs (k) , (A1)

respectively.
It is convenient to introduce a combined electrochemical potential Φ(r) + µ(r)
e , giving rise to a generalized (external)
force field defined by

µ(r)
E(r) = −∇r Φ(r) + , E = −∇r Φ , (A2)
e

where Φ(r) is the scalar potential. The Hamilton’s equations of motion for the quasiparticles, under the influence of a
static electrochemical potential and a static magnetic (B) field, are given by [68, 77, 78]

ṙ = ∇k ξs − k̇ × Ωs and k̇ = − e (E + ṙ × B)
⇒ ṙ = Ds [ws + e (E × Ωs ) + e (Ωs · ws ) B] and k̇ = − e Ds [E + (ws × B) + e (E · B) Ωs ] . (A3)

where − e is the charge carried by each quasiparticle. Furthermore,

1
Ds = (A4)
1 + e (B · Ωs )

is the factor which modifies the phase volume element from dVp to (Ds )−1 dVp , such that the Liouville’s theorem (in the
absence of collisions) continues to hold in the presence of a nonzero BC [79–82]. Putting everything together, the BE for
the quasiparticles takes the form [21, 83]

Ds ∂t + {ws + e E × Ωs + e (Ωs · ws ) B} · ∇r − e (E + ws × B) · ∇k − e2 (E · B) Ωs · ∇k fs = Icoll .

(A5)

which results from the Liouville’s equation in the presence of scattering events. On the right-hand side, Icoll denotes the
collision integral, which corrects the Liouville’s equation, taking into account the collisions of the quasiparticles. We would
like to point out that the term represented by ∇r µ was missed in Ref. [61], which demonstrated the quantized ECR in
Weyl semimetals.
Let the contributions to the average DC electric and thermal current densities from the quasiparticles, associated with
the band s at the node with chirality χ, be Js and Jth s , respectively. The response matrix, which relates the resulting
generalized current densities to the driving electric potential gradient and temperature gradient, is expressed as
" # " #" #
(Js )a X (σs )ab (αs )ab Eb
= , (A6)
Jsth a b
T (αs )ab (ℓs )ab − ∂b T

where {a, b} ∈ {x, y, z} indicates the Cartesian components of the current density vectors and the response tensors in 3d.
The symbols σs and αs represents the magnetoelectric conductivity and the magnetothermoelectric conductivity tensors,
respectively. The latter determines the Peltier (Πs ), Seebeck (Ss ), and Nernst coefficients. The third tensor ℓ represents
the response relating the thermal current density to the temperature gradient, at a vanishing electric field. S , Π, and the
magnetothermal coefficient tensor κ (which provides the coefficients between the heat current density and the temperature
gradient at vanishing electric current) are related as [34, 68]:
−1 −1 −1
X X X
(Ss )ab = (σs )aa′ (αs )a′ b , (Πs )ab = T (αs )aa′ (σs )a′ b , (κs )ab = (ℓs )ab − T (αs )aa′ (σs )a′ b′ (αs )b′ b . (A7)
i′ a′ a′ , b′

Since ℓs determines the first term in the magnetothermal coefficient tensor κs , here we will loosely refer to ℓs itself as the
magnetothermal coefficient.
The explicit form of the dc charge current current density is given by [67, 84, 85]

d3 k d3 k
Z Z
−1
Jtot
s = − e gs Ds ṙ f s (r, k) + ∇ r × Ms (r) , Ms (r) = gs D−1 ms (k) fs (r, k) , (A8)
(2 π) 3 (2 π)3 s

where Ms (r) is the magnetization density. As discussed earlier, ms (k) represents the the orbital magnetic moment,
which generically describes the rotation of a wavepacket around its center of mass. The contribution from the magnetic
moments to the local current density must be subtracted out in the transport current, because the magnetization current
11

cannot be measured by conventional transport experiments [86]. Therefore, the transport current is defined by Js =
R d3 k −1
Jtot
s − gs ∇r × (2 π)3 Ds ms (k) fs (r, k), leading to

d3 k
Z
Js = − e gs D−1 ṙ fs (r, k) . (A9)
(2 π)3 s

In an analogous way, the transport heat current is captured by [84]

d3 k
Z
Jth
s = gs (Ds )−1 ṙ (ξs − µ) fs (r, k) . (A10)
(2 π)3

Plugging in the expressions shown in in Eq. (A3), we get

d3 k
Z
Js = − e gs [ws + e (E × Ωs ) + e (Ωs · ws ) B] fs (r, k)
(2 π)3
d3 k
Z
and Jth
s = gs [ws + e (E × Ωs ) + e (Ωs · ws ) B] (ξs − µ) fs (r, k) . (A11)
(2 π)3

Under the relaxation-time approximation, the collision integral takes the form of
(0)
fs (r, k) − fs (r, k, t)
Icoll = , (A12)
τ
where the time-independent distribution function
1
fs(0) (r, k) ≡ f0 ξs (k), µ, T (r) =

h i, (A13)
ξs (k)−µ
1 + exp T (r)

describes a local equilibrium situation at the subsystem centred at position r, at the local temperature T (r), and with a
local chemical potential µ. The gradients of the equilibrium distribution function evaluate to

∇r T
∇k f0 (ξs ) = ws f0′ (ξs ) and ∇r f0 (ξs ) = − ∇r µ + (ξs − µ) f0′ (ξs ), (A14)
T

where the“prime” superscript is used to indicate partial-differentiation with respect to the variable shown within the
brackets [for example, f0′ (u) ≡ ∂u f0 (u)]. Henceforth, we set gs = 1, ignoring the degeneracy due to electron’s spin.
In order to obtain a solution to the full BE, for small time-independent values of E, ∇r T , and ∇r µ, we assume a small
deviation from the equilibrium distribution of the quasiparticles, such that

fs (r, k, t) ≡ fs (r, k) = f0 + δfs (r, k) . (A15)

Here, we have not included any explicit time-dependence in δfs (r, k) because the applied fields and gradients are static.
Furthermore, we have suppressed showing explicitly the dependence of f0 on ξs (k), µ, and T (r). We now parametrize the
∞
(p)
ϵp fs , where ϵ ∈ [0, 1] is the perturbative parameter having the same order of smallness as
P
deviation as δfs (r, k) =
p=1
the external perturbations E, ∇r T , and ∇r µ. Since ϵ is used solely for bookkeeping purpose (to track the order in the
degree of smallness), we will set ϵ = 1 at the end of the calculations after we solve for δfs (r, k) recursively, order by order
in increasing powers of ϵ.

Appendix B: Useful integrals

In the main text, we have to deal with integrals of the form:

d3 k
Z
I= F (k, εs ) f0′ (εs ) , (B1)
(2π)3

where εs = s v0 k + v0 η kz . We switch to the spherical polar coordinates such that

ε̃ cos ϕ sin θ ε̃ sin ϕ sin θ ε̃ cos θ

kx = , ky = , kz = , (B2)
s v0 s v0 s v0
12

ε̃2 sin θ
where ε̃ ∈ [0, ∞), ϕ ∈ [0, 2π), and θ ∈ [0, π]. The Jacobian of the transformation is J (ε̃, θ) = s3 v03
. This leads to
Z ∞ Z ∞ Z 2π Z π
η cos θ
d3 k → dε̃ dϕ dθ J (ε̃, θ) and εs (k) → εs (ε̃, θ) = ε̃ Λs (θ), with Λs (θ) = 1 + . (B3)
−∞ 0 0 0 s
Since we have chosen the tilting direction with respect to the z-axis, the dispersion does not depend on ϕ. Hence, we
can perform the ϕ-integration easily, after which I can be written in the following schematic form:
Z ∞ Z π
I= dε dθ I1 (ε, θ) f0′ (εs ) . (B4)
0 0

where I1 (ε, θ) is the function obtained after the ϕ-integration.

In our computations, the ε̃- and θ-dependent parts of the integrand are decoupled. Hence, let us discuss
R∞ some generic
identities for these two kinds of integrals. For the ε̃-integration, we encounter integrals of the forms 0 dε̃ ε̃n f0′ (εs ) and
R∞ λ+1

0
dε̃ ε̃n ∂ ∂ελ+1
f0 (εs )
, where {n, λ} ∈ Z. Applying the Sommerfeld expansion [68], we get
s

∞ ∞ ∞
β eβ(εs −µ) β eβ (ε̃−µ)
Z Z Z
1 Υn (µ, T )
dε̃ ε̃n f0′ (εs ) = − dε̃ ε̃n 2 = − n+1 dε̃ ε̃n β (ε̃−µ) 2
= − n+1 ,
0 0 1 + eβ (εs −µ) Λs (θ) 0 (1 + e ) Λs (θ)
(B5)
where
π 2 T 2 n (n − 1) 7 π 2 T 4 n (n − 3) (n − 2) (n − 1)

µ −6
Υn (µ, T ) = µn 1 + 2
+ 4
+ O , (B6)
6µ 360 µ T
which is valid in the regime β µ ≫ 1 (or µ ≫ T in the natural units). It is easy to show that [35, 36], for higher-order
derivatives, we have the relation
Z ∞
∂ λ+1 f0 (ε̃) n! Υn−λ (µ, T )
dε̃ ε̃n (−1)λ+1 = . (B7)
0 ∂ ε̃ λ+1 (n − λ)! Λn+1
s (θ)
For the θ-integration, we use the identity

π m n Z 1 m−1
1 − t2 2 tn
Z
(sin θ) (cos θ)
dθ = dt
0 (s + η cos θ)l −1 (s + η t)l
√ "
π Γ( m+1
2 ) n
n + 1 n + 1 l + 1 l 1 1 η2
= 2 {(−1) + 1} Γ 3 F̃2 , , ; , (m + n + 2);
4 sl 2 2 2 2 2 2 s2
#
ηl n n + 2 l + 1 l + 2 3 m + n + 3 η2
n
+ {(−1) − 1} Γ + 1 3 F̃2 , , ; , ; 2 , (B8)
s 2 2 2 2 2 2 s

where n1 F̃n2 {a1 , . . . , an1 }; {b1 , . . . , bn2 }; X is the regularized hypergeometric function [87].

Appendix C: Linear-in-B parts of magnetoelectric conductivity

In this appendix, we outline the forms of the various components of the magnetoelectric conductivity tensor, involving
the linear-in-B parts, for the planar Hall set-up with E and B aligned parallel to each other (discussed in Sec. IV).

1. Intrinsic anomalous-Hall part

From the term proportional to (E × Ωs ) in the integrand of Eq. (A11), we get the electric current density as

d3 k
Z
AH 2
Js = − e [(E × Ωs )] f0 (ξs ) , (C1)
(2 π)3

which gives the intrinsic anomalous-Hall term. Using f0 (ξs ) = f0 (εs ) + ζs f0′ (εs ) + O B 2 , we get

d3 k
Z
AH 2
(Ωs )c f0 (εs ) + ζs f0′ (εs ) + O B 2 ,

(σs )ab = − e ϵabc (C2)
(2 π)3
13

whose diagonal components are automatically zero because of the Levi-Civita symbol. A nonzero OMM may generate
B-dependent terms in (σsAH )ab .
For η = 0, the contribution from the first term vanishes identically. For η ̸= 0, although we get nonzero components from
this first term, they are B-independent and does not contribute to ςas . Let us consider the second term in the integrand,
which is proportional to ζs f0′ (εs ). First let us assume that êE = x̂, which involves setting ζs = e χ v0 Gs kx B/k 2 . The
corresponding contribution to
1. (σsAH )yx is

d3 k
Z
kx
tyx ≡ − e3 χ v0 Gs B ϵyxz (Ωs )z 2 f0′ (εs ) = 0 ; (C3)
(2 π)3 k

2. (σsAH )zx is
d3 k
Z
kx
tzx ≡ − e3 χ v0 Gs B ϵzxy (Ωs )y 2 f0′ (εs ) = 0 . (C4)
(2 π)3 k

Invoking the rotation symmetry of the dispersion about the kx ky -plane, we infer that, for êE = ŷ, we have txy = tzy = 0.
Next, let us assume that êE = ẑ, which involves setting ζs = e χ v0 Gs kz B/k 2 . The corresponding contribution to
1. (σsAH )xz is
d3 k
Z
kz
3
txz ≡ − e χ v0 Gs B ϵxzy 3
(Ωs )y 2 f0′ (εs ) = 0 ; (C5)
(2 π) k

2. (σsAH )yz is

d3 k
Z
kz
tyz ≡ − e3 χ v0 Gs B ϵyzx (Ωs )x 2 f0′ (εs ) = 0 . (C6)
(2 π)3 k

2. Lorentz-force contribution

The leading-order contribution from the Lorentz-force part is obtained by picking up the n = 1 term in the expression
for gs [shown in Eq. (34)], i.e., by using

δfsχ (k) = e2 τ 2 Ds f0′ (ξs ) L̂ [Ds (ws + Ws ) · E] . (C7)

This leads to the conductivity [36]
d3 k Ds2 2 Bc kd f0′ (ξs )
Z
(σsLF )ab = − ϵbdc e3 τ 2 s3 v03 [(w s ) a + (W s ) a ] ε 2
s − λ s ,
(2 π)3 ε5s k
X
λs = 2 χ e s Gs v02 ϱa′ Ba′ , ϱ = cos ϕ sin θ x̂ + sin ϕ sin θ ŷ + cos θ ẑ , (C8)
a′

where θ and ϕ refer to the polar and azimuthal angles used in the spherical polar coordinate transformation in Eq. (B2).
Using the same arguments as for the intrinsic anomalous-Hall part, we find that all the linear-in-B components of (σsLF )ab
vanish.

3. Non-anomalous-Hall contribution

The non-anomalous-Hall part of the current is given by [cf. Eq. (A11)]

d3 k
Z
J̄s = − e2 τ [(ws + Ws ) B] Ds [(ws + Ws ) · E] f0′ (ξs ) , (C9)
(2 π)3
leading to
d3 k
Z
(σ̄s )ab = − e2 τ Ds [(ws )a + (Ws )a ] [(ws )b + (Ws )b ] f0′ (ξs ) . (C10)
(2 π)3

We want to compute here the linear-in-B part of σ̄s , after dividing it up as σsBC + σsm , where σsBC arises solely due to the
effect of the BC and survives when OMM is set to zero, and σsm is the one which goes to zero if OMM is ignored.
14

a. BC-only part (no OMM)

The BC-only part is given by

e3 τ
Z
(vs )a (vs )b
(σsBC )ab = − d3 k Mab f0′ (ξs ) , Mab = − (Ωs · B) + (vs )a (vs · Ωs ) Bb + a ↔ b . (C11)
(2 π)3 2

b. Part with the integrand proportional to nonzero powers of OMM

In order to actually carry out the integration for this part, it is convenient to express us as
 
X ∆11 ∆12 ∆13
(us )a = (Us )ab Bb , Us = ∆21 ∆22 ∆23  , (C12)
b ∆31 ∆32 ∆33
where
sin2 θ cos(2ϕ) − cos2 θ π
∆11 (ε̃, θ, ϕ) = − χ e Gs s2 v03 , − ϕ) ,
∆22 (ε̃, θ, ϕ) = ∆11 (ε̃, θ,
ε̃2 2
sin2 θ sin(2ϕ)
∆33 (ε̃, θ, ϕ) = − ∆11 (ε̃, θ, 0) , ∆12 (ε̃, θ, ϕ) = ∆21 (ε̃, θ, ϕ) = − χ e Gs s2 v03 ,
ε̃2
∆12 ∆12
∆13 (ε̃, θ, ϕ) = ∆31 (ε̃, θ, ϕ) = , ∆23 (ε̃, θ, ϕ) = ∆32 (ε̃, θ, ϕ) = . (C13)
tan θ sin ϕ tan θ cos ϕ
We can now express the relevant part of conductivity as
d3 k
Z
(σsm )ab = − e2 τ [Sab f0′ (ξs ) + Pab f0′′ (ξs )] ,
(2 π)3
" #
X
c (vs )a (vs )b
Sab = (vs )a (Us )bc B + a ↔ b , Pab = − (ms · B) + a ↔ b . (C14)
c
2

4. Final expressions

Expressing the linear-in-B part of the current and the conductivity as Jalin,s = σslin ab
E b and σslin ab
, respectively, let
us define a third-rank tensor

lin c
∂ σslin ab
ςs ab
= . (C15)
∂Bc
For our scenario with E ∥ B, we will be dealing with the vector defined as
X c X b
ςas = ςslin ab δ b c = ςslin ab . (C16)
b,c b

Gathering all the ingredients from the preceding subsections, we divide up ςas as
ςas = ςa(s,1) + ςa(s,2) + ςa(s,3) + ςa(s,4) + ςa(s,5) , (C17)
where
XZ d3 k d3 k
Z
b ′
ςa(s,1) =e τ3
(v s ) a (v s ) (Ω s ) b f 0 (ε s ) = e 3
τ (vs )a (vs · Ωs ) f0′ (εs ) ,
(2 π)3 (2 π)3
b
d3 k
X Z
ςa(s,2) 3
= −e τ (1 + δab ) (vs )a (vs · Ωs ) f0′ (εs ) = −4 ςa(s,1) ,
(2 π)3
b
X Z d3 k e3 τ v0 Gs
Z
d3 k e v0 Gs k
′
ςa(s,3) 2
=e τ (v s ) a [∂k b
(m s ) b ] f 0 (ε s ) = (vs )a [∇k · (k Ωs )] f0′ (εs ) [since ms (k) = Ωs (k)] ,
(2 π)3 s (2 π)3 s
b
X Z d3 k b e3 τ v0 Gs
Z
d3 k
′
ςa(s,4) 2
=e τ (v s ) [∂ k a (m s ) b ] f 0 (ε s ) = [vs · ∂ka (k Ωs )] f0′ (εs ) ,
(2 π)3 s (2 π)3
b
X Z d3 k b e3 τ v0 Gs
Z
d3 k
′′
ςa(s,5) 2
=e τ 3
(v )
s a (v s ) (m )
s b 0 f (ε s ) = (vs )a k (vs · Ωs ) f0′′ (εs ) . (C18)
(2 π) s (2 π)3
b
15

(s,1) (s,2) (s,3)

While ςa and ςa represent the parts arising purely from the BC (i.e., with no contribution from the OMM), ςa ,
(s,4) (s,5)
ςa , and ςa are the terms which go to zero if OMM is neglected.
Below, we elucidate the final expressions:
1. Noting that vs · Ωs = − χ kv20 s (s + η cos θ), and using Eq. (B5), we obtain

d3 k δza η v0 e3 τ Cs
Z
(s,1) 3 s + η cos θ ′
ςa = − δza χ v0 s e τ (η + s cos θ) f 0 (ε s ) = − . (C19)
(2 π)3 k2 4 π2

2. Noting that ∇k · (k Ωs ) = − χ s/k 2 , and using Eq. (B5), we obtain

δza χ e3 τ v02 Gs
Z
η + s cos θ ′
ςa(s,3) = − d3 k f0 (εs )
(2 π)3 k2
δza χ e3 τ v02 Gs ε̃2 sin θ η + s cos θ s f0′ (ε̃)
Z Z
=− 2
dθ dε̃ 3 3
(2 π) s v0 k2 s + η cos θ
−1 η
2 2

3
δza χ v0 e τ Gs s − η tanh s −ηs
=− for |η| < |s| . (C20)
2 π2 η2

3. Noting that vs · ∂ka (k Ωs ) = − χ s (vs )a /k 2 , we obtain

δza χ s e3 τ v0 Gs d3 k (vs )z ′
Z
ςa(s,4) = − f (εs ) = ςa(s,3) . (C21)
s (2 π)3 k 2 0

4. Noting that vs · Ωs = − χ kv20 s (s + η cos θ), we obtain

d3 k
Z
(s,5) 3 3 s + η cos θ ′′
ςa = − δza χ e τ v0 Gs 3
(η + s cos θ) f0 (εs ) = − ςa(s,3) . (C22)
(2 π) k

[1] A. A. Burkov and L. Balents, Weyl semimetal in a topological insulator multilayer, Phys. Rev. Lett. 107, 127205 (2011).
[2] B. Yan and C. Felser, Topological materials: Weyl semimetals, Annual Rev. of Condensed Matter Phys. 8, 337 (2017).
[3] N. P. Armitage, E. J. Mele, and A. Vishwanath, Weyl and Dirac semimetals in three-dimensional solids, Rev. Mod. Phys. 90,
015001 (2018).
[4] M. M. H. Polash, S. Yalameha, H. Zhou, K. Ahadi, Z. Nourbakhsh, and D. Vashaee, Topological quantum matter to topological
phase conversion: Fundamentals, materials, physical systems for phase conversions, and device applications, Materials Science
and Engineering: R: Reports 145, 100620 (2021).
[5] S. Sekh and I. Mandal, Circular dichroism as a probe for topology in three-dimensional semimetals, Phys. Rev. B 105, 235403
(2022).
[6] I. Mandal, Signatures of two- and three-dimensional semimetals from circular dichroism, International Journal of Modern
Physics B 38, 2450216 (2024).
[7] J. E. Moore, Optical properties of Weyl semimetals, National Science Rev. 6, 206 (2018).
[8] C. Guo, V. S. Asadchy, B. Zhao, and S. Fan, Light control with Weyl semimetals, eLight 3, 2 (2023).
[9] A. Avdoshkin, V. Kozii, and J. E. Moore, Interactions remove the quantization of the chiral photocurrent at Weyl points, Phys.
Rev. Lett. 124, 196603 (2020).
[10] I. Mandal, Effect of interactions on the quantization of the chiral photocurrent for double-Weyl semimetals, Symmetry 12
(2020).
[11] M. Papaj and L. Fu, Magnus Hall effect, Phys. Rev. Lett. 123, 216802 (2019).
[12] D. Mandal, K. Das, and A. Agarwal, Magnus Nernst and thermal Hall effect, Phys. Rev. B 102, 205414 (2020).
[13] Sekh, Sajid and Mandal, Ipsita, Magnus Hall effect in three-dimensional topological semimetals, Eur. Phys. J. Plus 137, 736
(2022).
[14] F. D. M. Haldane, Berry curvature on the Fermi surface: Anomalous Hall effect as a topological Fermi-liquid property, Phys.
Rev. Lett. 93, 206602 (2004).
[15] P. Goswami and S. Tewari, Axionic field theory of (3 + 1)-dimensional Weyl semimetals, Phys. Rev. B 88, 245107 (2013).
[16] A. A. Burkov, Anomalous Hall effect in Weyl metals, Phys. Rev. Lett. 113, 187202 (2014).
[17] S.-B. Zhang, H.-Z. Lu, and S.-Q. Shen, Linear magnetoconductivity in an intrinsic topological Weyl semimetal, New Journal
of Phys. 18, 053039 (2016).
[18] Q. Chen and G. A. Fiete, Thermoelectric transport in double-Weyl semimetals, Phys. Rev. B 93, 155125 (2016).
[19] S. Nandy, G. Sharma, A. Taraphder, and S. Tewari, Chiral anomaly as the origin of the planar Hall effect in Weyl semimetals,
Phys. Rev. Lett. 119, 176804 (2017).
[20] S. Nandy, A. Taraphder, and S. Tewari, Berry phase theory of planar Hall effect in topological insulators, Scientific Reports 8,
14983 (2018).
16

[21] K. Das and A. Agarwal, Linear magnetochiral transport in tilted type-I and type-II Weyl semimetals, Phys. Rev. B 99, 085405
(2019).
[22] K. Das and A. Agarwal, Thermal and gravitational chiral anomaly induced magneto-transport in Weyl semimetals, Phys. Rev.
Res. 2, 013088 (2020).
[23] S. Das, K. Das, and A. Agarwal, Nonlinear magnetoconductivity in Weyl and multi-Weyl semimetals in quantizing magnetic
field, Phys. Rev. B 105, 235408 (2022).
[24] O. Pal, B. Dey, and T. K. Ghosh, Berry curvature induced magnetotransport in 3D noncentrosymmetric metals, Journal of
Phys.: Condensed Matter 34, 025702 (2022).
[25] O. Pal, B. Dey, and T. K. Ghosh, Berry curvature induced anisotropic magnetotransport in a quadratic triple-component
fermionic system, Journal of Phys.: Condensed Matter 34, 155702 (2022).
[26] L. X. Fu and C. M. Wang, Thermoelectric transport of multi-Weyl semimetals in the quantum limit, Phys. Rev. B 105, 035201
(2022).
[27] Y. Araki, Magnetic Textures and Dynamics in Magnetic Weyl Semimetals, Annalen der Physik 532, 1900287 (2020).
[28] Y. P. Mizuta and F. Ishii, Contribution of Berry curvature to thermoelectric effects, Proceedings of the International Conference
on Strongly Correlated Electron Systems (SCES2013), JPS Conf. Proc. 3, 017035 (2014).
[29] S. Yadav, S. Fazzini, and I. Mandal, Magneto-transport signatures in periodically-driven Weyl and multi-Weyl semimetals,
Physica E Low-Dimensional Systems and Nanostructures 144, 115444 (2022).
[30] A. Knoll, C. Timm, and T. Meng, Negative longitudinal magnetoconductance at weak fields in Weyl semimetals, Phys. Rev. B
101, 201402 (2020).
[31] L. Medel Onofre and A. Martı́n-Ruiz, Planar Hall effect in Weyl semimetals induced by pseudoelectromagnetic fields, Phys.
Rev. B 108, 155132 (2023).
[32] R. Ghosh and I. Mandal, Electric and thermoelectric response for Weyl and multi-Weyl semimetals in planar Hall configurations
including the effects of strain, Physica E: Low-dimensional Systems and Nanostructures 159, 115914 (2024).
[33] R. Ghosh and I. Mandal, Direction-dependent conductivity in planar Hall set-ups with tilted Weyl/multi-Weyl semimetals,
Journal of Physics: Condensed Matter 36, 275501 (2024).
[34] I. Mandal and K. Saha, Thermoelectric response in nodal-point semimetals, Annalen der Physik 536, 2400016 (2024).
[35] L. Medel, R. Ghosh, A. Martı́n-Ruiz, and I. Mandal, Electric, thermal, and thermoelectric magnetoconductivity for Weyl/multi-
Weyl semimetals in planar Hall set-ups induced by the combined effects of topology and strain, Scientific Reports 14, 21390
(2024).
[36] R. Ghosh, F. Haidar, and I. Mandal, Linear response in planar Hall and thermal Hall setups for Rarita-Schwinger-Weyl
semimetals, Phys. Rev. B 110, 245113 (2024).
[37] V. Gusynin, S. Sharapov, and J. Carbotte, Magneto-optical conductivity in graphene, Journal of Phys.: Condensed Matter 19,
026222 (2006).
[38] M. Stålhammar, J. Larana-Aragon, J. Knolle, and E. J. Bergholtz, Magneto-optical conductivity in generic Weyl semimetals,
Phys. Rev. B 102, 235134 (2020).
[39] S. Yadav, S. Sekh, and I. Mandal, Magneto-optical conductivity in the type-I and type-II phases of Weyl/multi-Weyl semimetals,
Physica B: Condensed Matter 656, 414765 (2023).
[40] I. Mandal and A. Sen, Tunneling of multi-Weyl semimetals through a potential barrier under the influence of magnetic fields,
Phys. Lett. A 399, 127293 (2021).
[41] S. Bera and I. Mandal, Floquet scattering of quadratic band-touching semimetals through a time-periodic potential well, Journal
of Phys. Condensed Matter 33, 295502 (2021).
[42] S. Bera, S. Sekh, and I. Mandal, Floquet transmission in Weyl/multi-Weyl and nodal-line semimetals through a time-periodic
potential well, Ann. Phys. (Berlin) 535, 2200460 (2023).
[43] D. Sinha and K. Sengupta, Transport across junctions of a weyl and a multi-weyl semimetal, Phys. Rev. B 99, 075153 (2019).
[44] I. Mandal, Transmission and conductance across junctions of isotropic and anisotropic three-dimensional semimetals, European
Physical Journal Plus 138, 1039 (2023).
[45] B. Bradlyn, J. Cano, Z. Wang, M. G. Vergniory, C. Felser, R. J. Cava, and B. A. Bernevig, Beyond Dirac and Weyl fermions:
Unconventional quasiparticles in conventional crystals, Science 353 (2016).
[46] L. Liang and Y. Yu, Semimetal with both Rarita-Schwinger-Weyl and Weyl excitations, Phys. Rev. B 93, 045113 (2016).
[47] I. Boettcher, Interplay of topology and electron-electron interactions in Rarita-Schwinger-Weyl semimetals, Phys. Rev. Lett.
124, 127602 (2020).
[48] J. M. Link, I. Boettcher, and I. F. Herbut, d-wave superconductivity and Bogoliubov-Fermi surfaces in Rarita-Schwinger-Weyl
semimetals, Phys. Rev. B 101, 184503 (2020).
[49] H. Isobe and L. Fu, Quantum critical points of j = 23 Dirac electrons in antiperovskite topological crystalline insulators, Phys.
Rev. B 93, 241113 (2016).
[50] P. Tang, Q. Zhou, and S.-C. Zhang, Multiple types of topological fermions in transition metal silicides, Phys. Rev. Lett. 119,
206402 (2017).
[51] 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).
[52] J.-Z. Ma, Q.-S. Wu, M. Song, S.-N. Zhang, E. Guedes, S. Ekahana, M. Krivenkov, M. Yao, S.-Y. Gao, W.-H. Fan, et al.,
Observation of a singular Weyl point surrounded by charged nodal walls in ptga, Nature Communications 12, 3994 (2021).
[53] I. Mandal, Andreev bound states in superconductor-barrier-superconductor junctions of Rarita-Schwinger-Weyl semimetals,
Physics Letters A 503, 129410 (2024).
[54] S. Weinberg, The quantum theory of fields. Vol. 3: Supersymmetry (Cambridge University Press, 2013).
[55] J. Cayssol and J. N. Fuchs, Topological and geometrical aspects of band theory, Journal of Physics: Materials 4, 034007 (2021).
[56] H. Nielsen and M. Ninomiya, A no-go theorem for regularizing chiral fermions, Phys. Lett. B 105, 219 (1981).
[57] D. Takane, Z. Wang, S. Souma, K. Nakayama, T. Nakamura, H. Oinuma, Y. Nakata, H. Iwasawa, C. Cacho, T. Kim, K. Horiba,
17

H. Kumigashira, T. Takahashi, Y. Ando, and T. Sato, Observation of chiral fermions with a large topological charge and
associated Fermi-arc surface states in CoSi, Phys. Rev. Lett. 122, 076402 (2019).
[58] D. S. Sanchez, I. Belopolski, T. A. Cochran, X. Xu, J.-X. Yin, G. Chang, W. Xie, K. Manna, V. Süß, C.-Y. Huang, et al.,
Topological chiral crystals with helicoid-arc quantum states, Nature 567, 500 (2019).
[59] N. B. Schröter, D. Pei, M. G. Vergniory, Y. Sun, K. Manna, F. De Juan, J. A. Krieger, V. Süss, M. Schmidt, P. Dudin, et al.,
Chiral topological semimetal with multifold band crossings and long fermi arcs, Nature Physics 15, 759 (2019).
[60] B. Q. Lv, Z.-L. Feng, J.-Z. Zhao, N. F. Q. Yuan, A. Zong, K. F. Luo, R. Yu, Y.-B. Huang, V. N. Strocov, A. Chikina, A. A.
Soluyanov, N. Gedik, Y.-G. Shi, T. Qian, and H. Ding, Observation of multiple types of topological fermions in pdbise, Phys.
Rev. B 99, 241104 (2019).
[61] R. Flores-Calderón and A. Martı́n-Ruiz, Quantized electrochemical transport in Weyl semimetals, Phys. Rev. B 103, 035102
(2021).
[62] F. Balduini, A. Molinari, L. Rocchino, V. Hasse, C. Felser, M. Sousa, C. Zota, H. Schmid, A. G. Grushin, and B. Gotsmann,
Intrinsic negative magnetoresistance from the chiral anomaly of multifold fermions, Nature Communications 15, 6526 (2024).
[63] F. Flicker, F. de Juan, B. Bradlyn, T. Morimoto, M. G. Vergniory, and A. G. Grushin, Chiral optical response of multifold
fermions, Phys. Rev. B 98, 155145 (2018).
[64] Y. Shen, Y. Jin, Y. Ge, M. Chen, and Z. Zhu, Chiral topological metals with multiple types of quasiparticle fermions and large
spin Hall effect in the SrGePt family materials, Phys. Rev. B 108, 035428 (2023).
[65] G. Chang, S.-Y. Xu, B. J. Wieder, D. S. Sanchez, S.-M. Huang, I. Belopolski, T.-R. Chang, S. Zhang, A. Bansil, H. Lin, and
M. Z. Hasan, Unconventional chiral fermions and large topological Fermi arcs in RhSi, Phys. Rev. Lett. 119, 206401 (2017).
[66] K. Nakazawa, T. Yamaguchi, and A. Yamakage, Nonlinear charge and thermal transport properties induced by orbital magnetic
moment in chiral crystal cobalt monosilicide, arXiv e-prints (2024), arXiv:2409.08040 [cond-mat.str-el].
[67] D. Xiao, M.-C. Chang, and Q. Niu, Berry phase effects on electronic properties, Rev. Mod. Phys. 82, 1959 (2010).
[68] N. Ashcroft and N. Mermin, Solid State Physics (Cengage Learning, 2011).
[69] C. Zeng, S. Nandy, and S. Tewari, Fundamental relations for anomalous thermoelectric transport coefficients in the nonlinear
regime, Phys. Rev. Res. 2, 032066 (2020).
[70] L. Onsager, Reciprocal Relations in Irreversible Processes. I., Phys. Rev. 37, 405 (1931).
[71] H. B. G. Casimir, On Onsager’s principle of microscopic reversibility, Rev. Mod. Phys. 17, 343 (1945).
[72] P. Jacquod, R. S. Whitney, J. Meair, and M. Büttiker, Onsager relations in coupled electric, thermoelectric, and spin transport:
The tenfold way, Phys. Rev. B 86, 155118 (2012).
[73] L. Li, C. Cui, R.-W. Zhang, Z.-M. Yu, and Y. Yao, Planar Hall plateau in magnetic Weyl semimetals, arXiv e-prints (2024),
arXiv:2406.11273 [cond-mat.mes-hall].
[74] I. Mandal, Chiral anomaly and internode scatterings in multifold semimetals, arXiv e-prints 10.48550/arXiv.2411.18434,
arXiv:2411.18434 [cond-mat.mes-hall].
[75] I. Mandal and K. Saha, Thermopower in an anisotropic two-dimensional Weyl semimetal, Phys. Rev. B 101, 045101 (2020).
[76] J. Xiong, S. Kushwaha, J. Krizan, T. Liang, R. J. Cava, and N. P. Ong, Anomalous conductivity tensor in the Dirac semimetal
Na3 Bi, EPL (Europhysics Letters) 114, 27002 (2016).
[77] G. Sundaram and Q. Niu, Wave-packet dynamics in slowly perturbed crystals: Gradient corrections and Berry-phase effects,
Phys. Rev. B 59, 14915 (1999).
[78] L. Li, J. Cao, C. Cui, Z.-M. Yu, and Y. Yao, Planar Hall effect in topological Weyl and nodal-line semimetals, Phys. Rev. B
108, 085120 (2023).
[79] D. T. Son and B. Z. Spivak, Chiral anomaly and classical negative magnetoresistance of Weyl metals, Phys. Rev. B 88, 104412
(2013).
[80] D. Xiao, J. Shi, and Q. Niu, Berry Phase Correction to Electron Density of States in Solids, Phys. Rev. Lett. 95, 137204 (2005).
[81] C. Duval, Z. Horváth, P. A. Horvathy, L. Martina, and P. Stichel, Berry phase correction to electron density in solids and
“exotic” dynamics, Mod. Phys. Lett. B 20, 373 (2006).
[82] D. T. Son and N. Yamamoto, Berry curvature, triangle anomalies, and the chiral magnetic effect in Fermi liquids, Phys. Rev.
Lett. 109, 181602 (2012).
[83] R. Lundgren, P. Laurell, and G. A. Fiete, Thermoelectric properties of Weyl and Dirac semimetals, Phys. Rev. B 90, 165115
(2014).
[84] D. Xiao, Y. Yao, Z. Fang, and Q. Niu, Berry-phase effect in anomalous thermoelectric transport, Phys. Rev. Lett. 97, 026603
(2006).
[85] K. Das and A. Agarwal, Berry curvature induced thermopower in type-I and type-II Weyl semimetals, Phys. Rev. B 100,
085406 (2019).
[86] N. R. Cooper, B. I. Halperin, and I. M. Ruzin, Thermoelectric response of an interacting two-dimensional electron gas in a
quantizing magnetic field, Phys. Rev. B 55, 2344 (1997).
[87] E. W. Weisstein, Regularized hypergeometric function, From MathWorld–A Wolfram Web Resource (Wolfram Research, Inc.).