Pseudospin-1 Fermions

• Pseudospin-1 fermions are emergent quasiparticles described by a three-component Dirac–Weyl Hamiltonian, featuring two dispersive cones and a central flat band.
• They exhibit unique transport phenomena such as perfect collimation, super-Andreev reflection, and quantum interference effects sensitive to lattice geometry and disorder.
• Synthetic realizations in optical lattices and spin–orbit coupled systems extend their applicability in exploring topological phases, high-Chern-number states, and novel superconducting interfaces.

Pseudospin-1 fermions are emergent quasiparticles whose low-energy dynamics are governed by a three-component Dirac–Weyl or Weyl Hamiltonian acting on an internal pseudospin-$1$ space rather than on the physical spin of an elementary particle. In the canonical massless model, the spectrum consists of two linearly dispersing bands and a central flat band, $E_0(\mathbf{k})=0$ and $E_\pm(\mathbf{k})=\pm \hbar v_F |\mathbf{k}|$, so that three bands meet at a threefold degeneracy. This structure appears in condensed-matter systems such as dice and $\alpha$-$\mathcal T_3$ lattices, in multifold semimetals, and in synthetic platforms including optical lattices, where related threefold crossings are also described as Maxwell points (Chen et al., 2018, Zhu et al., 2017).

1. Canonical Hamiltonians and band structure

The standard two-dimensional pseudospin-1 Dirac–Weyl Hamiltonian is

$H(\mathbf{k})=\hbar v_F (S_x k_x + S_y k_y),$

while the three-dimensional Weyl analogue near a node of chirality $\chi=\pm 1$ is

$$H^\chi(\mathbf{k})=\chi\,\hbar v_F\,\mathbf{k}\!\cdot\!\mathbf{S}.$$

In both cases $\mathbf{S}=(S_x,S_y,S_z)$ is represented by $3\times 3$ spin-1 matrices, and diagonalization yields a three-band spectrum with a central flat band and two cones. In three dimensions, the two dispersive bands carry Chern numbers $E_0(\mathbf{k})=0$0, while the flat band is topologically trivial; this doubles the monopole charge relative to an ordinary Weyl node (Ahmad et al., 2024).

The most widely used lattice interpolation is the $E_0(\mathbf{k})=0$1-$E_0(\mathbf{k})=0$2 model, equivalently parameterized by $E_0(\mathbf{k})=0$3 or $E_0(\mathbf{k})=0$4. At $E_0(\mathbf{k})=0$5 it reduces to graphene plus an inert flat band, whereas at $E_0(\mathbf{k})=0$6 it reaches the dice limit, where the middle band is strictly dispersionless at zero energy. In the gapless limit, the spectrum is $E_0(\mathbf{k})=0$7 and $E_0(\mathbf{k})=0$8; with a mass term, the dispersive bands become $E_0(\mathbf{k})=0$9, while the location of the flat band depends on the mass structure. For the “$E_\pm(\mathbf{k})=\pm \hbar v_F |\mathbf{k}|$0-type” mass term the flat band sits at $E_\pm(\mathbf{k})=\pm \hbar v_F |\mathbf{k}|$1 or $E_\pm(\mathbf{k})=\pm \hbar v_F |\mathbf{k}|$2; for the “$E_\pm(\mathbf{k})=\pm \hbar v_F |\mathbf{k}|$3-type” mass term it remains at zero energy, centered in the gap $E_\pm(\mathbf{k})=\pm \hbar v_F |\mathbf{k}|$4 (Gorbar et al., 2018, Zeng et al., 2021).

A persistent source of confusion is the status of the flat band. In the idealized massless Dirac–Weyl model it is pinned at zero energy and intersects the cones at the Dirac point. In more general $E_\pm(\mathbf{k})=\pm \hbar v_F |\mathbf{k}|$5-$E_\pm(\mathbf{k})=\pm \hbar v_F |\mathbf{k}|$6 models, however, the flat branch can move to a band edge or be destroyed by a long-range potential. This distinction is essential for transport, impurity binding, and magnetotransport (Gorbar et al., 2018).

2. Microscopic realizations, lattice geometry, and confinement

Several microscopic settings realize pseudospin-1 fermions. In crystalline lattice models, the dice or $E_\pm(\mathbf{k})=\pm \hbar v_F |\mathbf{k}|$7 lattice and the Lieb lattice provide the canonical two-dimensional examples with two cones crossing a flat band. In the dice lattice, equal nearest-neighbor hoppings produce the Bloch Hamiltonian with three sublattices $E_\pm(\mathbf{k})=\pm \hbar v_F |\mathbf{k}|$8, $E_\pm(\mathbf{k})=\pm \hbar v_F |\mathbf{k}|$9, and $\alpha$0, and the continuum limit near each Dirac point reduces to the pseudospin-1 Dirac Hamiltonian. Oriekhov, Gorbar, and Gusynin showed that the flat zero-energy band is robust in ribbons with both armchair and zigzag terminations, even though the confined spectra depend strongly on boundary type (Oriekhov et al., 2018).

The boundary problem is atypical because current conservation does not require continuity of all spinor components. For integer spin-$\alpha$1 Weyl systems, continuity across a boundary involves only $\alpha$2 out of $\alpha$3 components; for pseudospin-$\alpha$4, only two components need be continuous. In one formulation for transport across a planar interface this means continuity of $\alpha$5 and $\alpha$6; in another, appropriate for the BdG treatment of massive pseudospin-1 fermions, the continuous quantities are $\alpha$7 and $\alpha$8 on each electron or hole block. This reduced matching condition underlies the unusual barrier and Andreev phenomena of the pseudospin-1 case (Nandy et al., 2019, Zeng et al., 2021).

Ribbon confinement also differs from graphene. Armchair ribbons display the same quantization structure as graphene except for the additional flat band, whereas zigzag dice ribbons admit no propagating edge states localized at the boundary. Certain zigzag terminations instead generate metallic bulk modes, including an $\alpha$9 branch $\mathcal T_3$0. A plausible implication is that “zigzag” in pseudospin-1 systems should not be identified automatically with graphene-like edge transport (Oriekhov et al., 2018).

Beyond solid-state lattices, Zhu et al. proposed two- and three-dimensional optical-lattice realizations in which three internal atomic states encode the pseudospin-$\mathcal T_3$1 space. Their models support Maxwell points with Berry phase $\mathcal T_3$2 in two dimensions and monopole charge $\mathcal T_3$3 in three dimensions, together with anomalous quantum Hall phases and two Fermi arcs connecting the threefold nodes (Zhu et al., 2017).

3. Ballistic transport, pumping, and superconducting interfaces

Barrier transport in pseudospin-1 systems displays phenomena without a direct pseudospin-$\mathcal T_3$4 analogue. In normal-metal–barrier–normal-metal junctions, a barrier tuned to $\mathcal T_3$5 acts as a perfect collimator: every mode with nonzero incidence angle is reflected with unit probability for any barrier width $\mathcal T_3$6, whereas normal incidence retains Klein tunneling. The differential conductance also obeys the exact symmetry

$\mathcal T_3$7

with the $\mathcal T_3$8 sign for particle transport and the $\mathcal T_3$9 sign for hole transport (Nandy et al., 2019).

In adiabatically driven double-barrier structures, the three-band spectrum causes sharp current-direction reversal near the Dirac point. The pumped current is governed by the Berry curvature of the instantaneous scattering matrix in the $H(\mathbf{k})=\hbar v_F (S_x k_x + S_y k_y),$0 parameter plane, and the sign reversal coincides with parameter regimes where a higher barrier can transmit more strongly than a lower barrier because of super–Klein tunneling or hole-like transmission below the Dirac point. Around $H(\mathbf{k})=\hbar v_F (S_x k_x + S_y k_y),$1, the angle-averaged pumped current shows sharp sign changes, while exactly at the Dirac point the ballistic transmission vanishes but the pumped current remains finite once the flat-band channel is included consistently in boundary matching (Chen et al., 2018).

Normal-metal–superconductor interfaces are especially distinctive. For massless pseudospin-1 fermions, the Andreev reflection amplitude in the subgap regime can be solved in closed form, and in the limit $H(\mathbf{k})=\hbar v_F (S_x k_x + S_y k_y),$2 with $H(\mathbf{k})=\hbar v_F (S_x k_x + S_y k_y),$3, one obtains $H(\mathbf{k})=\hbar v_F (S_x k_x + S_y k_y),$4 for all incident angles. The paper terms this super-Andreev reflection: perfect transparency of the NS interface, a conductance peak with $H(\mathbf{k})=\hbar v_F (S_x k_x + S_y k_y),$5, and a direct transport signature near $H(\mathbf{k})=\hbar v_F (S_x k_x + S_y k_y),$6 when $H(\mathbf{k})=\hbar v_F (S_x k_x + S_y k_y),$7. The same analysis finds a sizable longitudinal Goos–Hänchen-type shift, with the Andreev-reflected hole shifted always in the forward direction in the subgap regime, irrespective of whether reflection is retro or specular (Feng et al., 2019).

Mass terms split this behavior into two classes. For the “$H(\mathbf{k})=\hbar v_F (S_x k_x + S_y k_y),$8-type” mass, Andreev reflection can be enhanced at oblique incidence relative to normal incidence. In retro reflection, that enhancement occurs in $H(\mathbf{k})=\hbar v_F (S_x k_x + S_y k_y),$9-doped $\chi=\pm 1$0-type and $\chi=\pm 1$1-doped $\chi=\pm 1$2-type systems; in specular reflection, it occurs in $\chi=\pm 1$3-doped $\chi=\pm 1$4-type and $\chi=\pm 1$5-doped $\chi=\pm 1$6-type systems. For the “$\chi=\pm 1$7-type” mass, an undoped junction at $\chi=\pm 1$8 and $\chi=\pm 1$9 exhibits $$H^\chi(\mathbf{k})=\chi\,\hbar v_F\,\mathbf{k}\!\cdot\!\mathbf{S}.$$0 for all $$H^\chi(\mathbf{k})=\chi\,\hbar v_F\,\mathbf{k}\!\cdot\!\mathbf{S}.$$1, again termed super Andreev reflection, now attributed to the flat band remaining at midgap (Zeng et al., 2021).

Magnetic barriers reverse the familiar comparison with pseudospin-$$H^\chi(\mathbf{k})=\chi\,\hbar v_F\,\mathbf{k}\!\cdot\!\mathbf{S}.$$2. For a $$H^\chi(\mathbf{k})=\chi\,\hbar v_F\,\mathbf{k}\!\cdot\!\mathbf{S}.$$3-function vector-potential barrier, pseudospin-$$H^\chi(\mathbf{k})=\chi\,\hbar v_F\,\mathbf{k}\!\cdot\!\mathbf{S}.$$4 Dirac fermions remain comparatively transparent, whereas pseudospin-$$H^\chi(\mathbf{k})=\chi\,\hbar v_F\,\mathbf{k}\!\cdot\!\mathbf{S}.$$5 transmission is strongly suppressed, allowing a pseudospin filter in which the pseudospin-$$H^\chi(\mathbf{k})=\chi\,\hbar v_F\,\mathbf{k}\!\cdot\!\mathbf{S}.$$6 channel is blocked while the pseudospin-$$H^\chi(\mathbf{k})=\chi\,\hbar v_F\,\mathbf{k}\!\cdot\!\mathbf{S}.$$7 channel still transmits (Li et al., 2024).

4. Magneto-optics, polarization response, and screening

The optical response of pseudospin-1 fermions reflects the coexistence of cones and a flat band. In a perpendicular magnetic field, Landau levels condense out of the three bands. Malcolm and Nicol obtained for the $$H^\chi(\mathbf{k})=\chi\,\hbar v_F\,\mathbf{k}\!\cdot\!\mathbf{S}.$$8 Dirac–Weyl model the dispersive levels

$$H^\chi(\mathbf{k})=\chi\,\hbar v_F\,\mathbf{k}\!\cdot\!\mathbf{S}.$$9

and a flat-band ladder

$\mathbf{S}=(S_x,S_y,S_z)$0

with $\mathbf{S}=(S_x,S_y,S_z)$1 missing. The optical selection rule is $\mathbf{S}=(S_x,S_y,S_z)$2, and flat-band transitions generate abrupt on/off spectral lines, unequal mirror-isosceles interband pairs, and polarization asymmetry between $\mathbf{S}=(S_x,S_y,S_z)$3 and $\mathbf{S}=(S_x,S_y,S_z)$4. These features provide a direct spectroscopic signature of the three-band structure (Malcolm et al., 2014).

The dynamical polarization function is likewise modified by flat-to-cone transitions. In the dice lattice, the random-phase-approximation dielectric function contains intraconal, interconal, and flat-to-cone contributions, producing a discontinuous step in $\mathbf{S}=(S_x,S_y,S_z)$5 at $\mathbf{S}=(S_x,S_y,S_z)$6 and a logarithmic divergence in $\mathbf{S}=(S_x,S_y,S_z)$7 as $\mathbf{S}=(S_x,S_y,S_z)$8. Malcolm and Nicol showed that this “pinches” the plasmon branch to the single point

$\mathbf{S}=(S_x,S_y,S_z)$9

independent of the background dielectric constant (Malcolm et al., 2016).

Static screening is also altered. The long-distance screened impurity potential has the same leading Thomas–Fermi decay as graphene, $3\times 3$0, but the Friedel term decays faster,

$3\times 3$1

rather than the graphene-like $3\times 3$2 envelope. This faster decay follows directly from the singular structure of the Lindhard function at $3\times 3$3 in the presence of the flat band (Malcolm et al., 2016).

5. Disorder, impurities, and localization

Disorder reveals perhaps the most counterintuitive property of pseudospin-1 fermions: the flat band does not itself carry current, yet it can strongly enhance transport anomalies. Vigh et al. analyzed the disordered dice lattice within self-consistent Born approximation and found that the sublattice-resolved density of states on the rim sites develops a narrow zero-energy peak of height $3\times 3$4, whereas the hub-site density of states remains small. At charge neutrality the dc conductivity behaves as

$3\times 3$5

so it diverges logarithmically as the disorder strength $3\times 3$6. A common misconception is that this divergence reflects transport by the flat band itself; Vigh et al. showed instead that it comes from interband transitions at the touching point between the flat band and the propagating cones (Vigh et al., 2013).

Impurity binding depends sensitively on the range of the potential. In the gapped $3\times 3$7-$3\times 3$8 model, bound-state levels descend from the upper and central continua and dive into the central and lower continua as the impurity strength increases. In the dice limit, the flat band survives in a compact-support potential well, but it is absent for the Coulomb potential. For the pure or regularized Coulomb problem, the wave functions admit power-series solutions, and the termination condition yields two quantization equations relating energy and charge, leading to a countably infinite set of quasi-exact bound states at special charges (Gorbar et al., 2018).

Quantum interference places pseudospin-1 fermions in the localization class of integer pseudospin. In the two-dimensional $3\times 3$9-$E_0(\mathbf{k})=0$00 model, quantum interference exhibits a crossover between weak antilocalization and weak localization as $E_0(\mathbf{k})=0$01 is tuned: graphene $E_0(\mathbf{k})=0$02 shows net WAL, a small finite $E_0(\mathbf{k})=0$03 enhances WAL because two Cooperon channels soften simultaneously, and the dice lattice yields maximal WL. In the exact arbitrary-pseudospin theory of two-dimensional fermions, the gapless Cooperon mode depends only on the pseudospin $E_0(\mathbf{k})=0$04, integer $E_0(\mathbf{k})=0$05 gives WL, and half-integer $E_0(\mathbf{k})=0$06 gives WAL (Singh et al., 2023, Singh et al., 2024).

The three-dimensional scalar-disorder theory reaches the same symmetry conclusion. For pseudospin-$E_0(\mathbf{k})=0$07, the system lies in the orthogonal class, and the leading quantum-interference correction takes the universal form

$E_0(\mathbf{k})=0$08

with the same magnitude as in conventional diffusive metals and Weyl fermions but with the sign fixed by the parity of $E_0(\mathbf{k})=0$09 (Gupta et al., 20 Apr 2026).

6. Berry curvature, anomalous transport, and the chiral anomaly

A mass term activates Berry curvature and orbital magnetic moment effects in the $E_0(\mathbf{k})=0$10-$E_0(\mathbf{k})=0$11 model. Singh and Sharma showed that in the dice limit the dispersive bands have

$E_0(\mathbf{k})=0$12

while $E_0(\mathbf{k})=0$13. Coupling of the orbital magnetic moment to an external field breaks valley symmetry, generates finite corrections to longitudinal and Hall conductivity, and produces an anomalous Hall conductivity due to the Berry curvature. The resulting magnetoresistance can be positive or negative, with opposite disorder trends for conventional and anomalous contributions (Singh et al., 2022).

In three-dimensional pseudospin-1 Weyl systems, the doubled monopole charge produces a doubled chiral-anomaly pumping rate. Ahmad and Sharma derived

$E_0(\mathbf{k})=0$14

with $E_0(\mathbf{k})=0$15 for each dispersive band. In the collinear geometry, the longitudinal magnetoconductance is positive and quadratic in $E_0(\mathbf{k})=0$16 for weak internode scattering, but changes sign beyond a critical intervalley ratio $E_0(\mathbf{k})=0$17–0.2; for Weyl fermions they reported $E_0(\mathbf{k})=0$18–0.3. They also found that the zero-field conductivity depends more strongly on internode scattering in the pseudospin-1 case, through a denominator $E_0(\mathbf{k})=0$19 rather than $E_0(\mathbf{k})=0$20 (Ahmad et al., 2024).

The planar Hall response extends this picture. A recent semiclassical Boltzmann treatment including momentum-dependent scattering, orbital magnetic moment corrections, and charge-conservation constraints found that in the untilted case the planar Hall conductance is positive and scales as $E_0(\mathbf{k})=0$21. Increasing the scattering strength induces a sign reversal, and a generic tilt changes the angular dependence to $E_0(\mathbf{k})=0$22 for tilt along $E_0(\mathbf{k})=0$23 or $E_0(\mathbf{k})=0$24 for tilt along $E_0(\mathbf{k})=0$25, together with a nonmonotonic dependence on tilt magnitude. Candidate materials identified for these signatures include space groups $E_0(\mathbf{k})=0$26, $E_0(\mathbf{k})=0$27, and $E_0(\mathbf{k})=0$28, with examples such as CoSi and PtGa (Ahmad, 9 Jun 2026).

7. Synthetic pseudospin-1 matter and topological phases

Optical lattices and spin-orbit-coupled cold-atom systems provide controlled routes to pseudospin-1 quasiparticles beyond the constraints of crystalline chemistry. In the optical-lattice proposal of Zhu et al., three internal atomic states on a square or cubic lattice realize low-energy Hamiltonians of the Maxwell form $E_0(\mathbf{k})=0$29. In two dimensions these systems support Maxwell points with Berry phase $E_0(\mathbf{k})=0$30, while in three dimensions each Maxwell point carries monopole charge $E_0(\mathbf{k})=0$31, and two Fermi arcs connect the nodes. For $E_0(\mathbf{k})=0$32 in two dimensions the bulk is a Chern insulator with lowest-band Chern number $E_0(\mathbf{k})=0$33, and the edge modes display spin-momentum locking (Zhu et al., 2017).

A distinct cold-atom route uses two-dimensional Dresselhaus-type spin-orbit coupling in spin-1 Fermi gases. The enlarged internal Hilbert space allows triply-degenerate points, quadratic band touching, and topological superfluids with large Chern numbers without relying on lattice point-group symmetries. In the paired phase, the BdG spectrum supports gapped states with $E_0(\mathbf{k})=0$34; the $E_0(\mathbf{k})=0$35 phase hosts five chiral Majorana edge modes, and upon adding a free $E_0(\mathbf{k})=0$36 dispersion the higher-order band touchings evolve into triple-Weyl nodes with monopole charge $E_0(\mathbf{k})=0$37 and cubic in-plane dispersion (Hou et al., 2018).

These synthetic constructions clarify a broader point. Pseudospin-1 fermions are not defined by a single lattice, material family, or dimensionality. What unifies the subject is the three-component low-energy structure, the coexistence of propagating cones and a flat or central band, and the way this internal geometry reshapes boundary conditions, transport, optical response, disorder physics, and topological classification across both solid-state and engineered quantum matter (Zhu et al., 2017, Hou et al., 2018).