Nematic Chiral d-Wave Superconductivity

• Nematic chiral d-wave superconductivity is characterized by a two-component d-wave order parameter on hexagonal/triangular lattices that distinguishes real, time-reversal-symmetric nematic states from complex, time-reversal-breaking chiral states.
• Microscopic selection mechanisms—including density-wave and gauge-field fluctuations, inter-band pairing, and frustrated multiband fermiology—lift the near-degeneracy between nematic and chiral orders.
• Practical insights stress that experimental probes such as STM, muSR, and angular transport measurements are vital for distinguishing nodal nematic patterns from fully gapped chiral states with topological edge features.

Nematic chiral d-wave superconductivity denotes the family of superconducting states generated by a two-component d-wave order parameter on lattices with hexagonal or triangular symmetry, where the components transforming as $d_{x^2-y^2}$ and $d_{xy}$ belong to a two-dimensional irreducible representation such as $E_2$, $E_g$, or $E_{2g}$. A real linear combination yields a nematic d-wave state that preserves time-reversal symmetry but lowers lattice rotational symmetry, whereas a complex combination such as $d_{x^2-y^2}\pm i d_{xy}$ yields a chiral state that breaks time-reversal symmetry and is often topological. In correlated systems the same manifold can also support states in which chirality and nematicity coexist because additional symmetry lowering, multiband structure, or extra pairing components deform the pure $d+id$ limit. A representative recent result is the half-filled isotropic triangular-lattice Hubbard model, where the nematic $d_{xy}$ state is the ground state below the Mott transition while chiral $d+id$ remains only slightly higher in energy and therefore quasi-stable (Yamada, 2024).

1. Symmetry structure and order-parameter manifold

On triangular, honeycomb, and related hexagonal lattices, the two d-wave harmonics form a two-component order parameter. In continuum notation these basis functions are commonly written as

$$\Delta_{d1}(\mathbf{k}) \propto k_x^2-k_y^2,\qquad \Delta_{d2}(\mathbf{k}) \propto 2k_xk_y,$$

so that the chiral combination

$d_{xy}$0

selects one of two opposite chiralities and breaks time-reversal symmetry (Fischer et al., 2013). On the isotropic triangular lattice the same pair of components is classified as the $d_{xy}$1 representation of $d_{xy}$2, and in the real-space construction used for the Hubbard model they appear as $d_{xy}$3 and $d_{xy}$4, with the chiral Weiss field taken as $d_{xy}$5 (Yamada, 2024).

The essential distinction between nematic and chiral states is symmetry rather than merely gap shape. A nematic d-wave state is a real, single-component or real two-component condensate; it preserves time-reversal symmetry and selects an orientation, lowering $d_{xy}$6 to a subgroup compatible with twofold anisotropy or retained mirror symmetries. A chiral $d_{xy}$7 state is complex; it breaks time-reversal symmetry and, on an ordinary Fermi surface, is typically fully gapped. The gap need not be fully isotropic, however: in SrPtAs the chiral $d_{xy}$8 state remains multiband-anisotropic and has point nodes only on one band, while in twisted bilayer graphene the chiral state is fully gapped but the nematic state is gapless with point nodes (Fischer et al., 2013, Wu, 2018).

A standard Ginzburg–Landau description expresses the two-component order parameter as $d_{xy}$9 and writes

$E_2$0

Within the convention used for honeycomb-lattice d-wave pairing, $E_2$1 favors the chiral solution $E_2$2, whereas $E_2$3 favors a real nematic solution (Black-Schaffer et al., 2014). Other papers use different coefficient parametrizations, but all describe the same competition inside the two-dimensional d-wave manifold.

2. Nematic, chiral, and mixed realizations

Pure nematic and pure chiral states are the two canonical condensates of the $E_2$4 manifold, but the literature also identifies mixed states in which chirality and nematicity coexist. On the isotropic triangular lattice, the nematic $E_2$5 state retains two mirror symmetries and places its line nodes at $E_2$6 relative to the axes of the reciprocal lattice vectors, whereas the chiral $E_2$7 state is fully gapped on the Fermi surface (Yamada, 2024). On the honeycomb lattice near the Mott state, the two real $E_2$8 bond patterns are

$E_2$9

and the chiral bond patterns are

$E_g$0

which realize the familiar $E_g$1 state (Black-Schaffer et al., 2014).

Moiré systems introduce an additional degree of freedom: the center-of-mass motion of Cooper pairs within the enlarged superlattice. In twisted bilayer graphene, the relative-motion pairing components $E_g$2 and $E_g$3 carry opposite angular momenta, while the center-of-mass envelope can itself carry angular momentum without breaking moiré periodicity. The resulting angular-momentum matching intrinsically couples opposite chiral relative-motion sectors and produces spontaneous vortex–antivortex textures in the center-of-mass component of one sector (Wu, 2018). This is not a simple restatement of the hexagonal $E_g$4 problem; it is a moiré-specific extension of it.

Several works go beyond the binary opposition between nematic and chiral. In kagome AV$E_g$5Sb$E_g$6, loop-current order can drive a chiral d-wave state, and the coexistence of loop-current order with Star-of-David charge-density-wave bond order produces a chiral state with pronounced nematicity even though the Fermi surfaces remain almost perfectly $E_g$7 symmetric (Tazai et al., 6 Aug 2025). On the honeycomb lattice, a tri-component order parameter

$E_g$8

spontaneously breaks both time-reversal and $E_g$9 rotational symmetry, leaving only $E_{2g}$0 and generating twelve degenerate domains; in that case nematicity does not replace chirality but deforms it through $E_{2g}$1–$E_{2g}$2 mixing (Li et al., 13 Apr 2025).

3. Microscopic selection mechanisms

The near-degeneracy between nematic and chiral d-wave order is rooted in representation theory. Because $E_{2g}$3 and $E_{2g}$4 form a two-dimensional irreducible representation, the singlet superconducting effective potential permits mixing between them. In the triangular-lattice Hubbard analysis, the two components are degenerate up to fourth order in a Landau expansion and sixth-order terms are required to lift the degeneracy, which explains why the numerically obtained energy splitting is small on the frustrated triangular lattice (Yamada, 2024). This is the basic reason the same symmetry manifold can host proximate nematic and chiral states.

Weak-coupling energetics frequently favor chirality because the complex combination removes nodes and gains condensation energy. The honeycomb-lattice $E_{2g}$5–$E_{2g}$6 and Hubbard analysis states that below $E_{2g}$7 a $E_{2g}$8 phase shift between the two $E_{2g}$9 components lowers the energy “very generally,” and RMFT plus DQMC find chiral $d_{x^2-y^2}\pm i d_{xy}$0 favored through a wide interaction range from intermediate to strong coupling (Black-Schaffer et al., 2014). A related weak-coupling Ginzburg–Landau treatment for hexagonal systems writes

$d_{x^2-y^2}\pm i d_{xy}$1

with $d_{x^2-y^2}\pm i d_{xy}$2 selecting chiral order and $d_{x^2-y^2}\pm i d_{xy}$3 selecting nematic order; within that framework, weak coupling yields $d_{x^2-y^2}\pm i d_{xy}$4, so strong density-wave fluctuations are required to reverse the sign and stabilize nematicity (Kozii et al., 2018).

Three distinct mechanisms for nematic stabilization recur in the current literature. First, density-wave fluctuations renormalize the quartic anisotropy and can make $d_{x^2-y^2}\pm i d_{xy}$5, thereby favoring a time-reversal-symmetric nematic d-wave state over the weak-coupling chiral state (Kozii et al., 2018). Second, electromagnetic gauge-field fluctuations generate a non-analytic cubic term after the transverse gauge modes are integrated out; because the superconducting stiffness is anisotropic, this cubic term depends on the relative phase and amplitudes of the two d-wave components and generally favors the nematic state over the chiral one, displacing chirality across a wide region of parameter space (Gali et al., 2022). Third, in multiple-flat-band systems without a conventional Fermi surface, inter-eigen-band pairing can dominate the condensation-energy balance. In the TBG heavy-fermion model, the nodal nematic Euler state has higher energy around the nodal points than the chiral state, but inter-eigen-band pairing lowers the energy over most of the moiré Brillouin zone, yielding a positive curvature coefficient $d_{x^2-y^2}\pm i d_{xy}$6 meV for representative parameters and making the nematic state energetically favorable overall (Liu et al., 2024).

Other mechanisms produce real, anisotropic mixtures without any chiral solution in the studied regime. In the partially flat-band triangular-lattice Hubbard model treated with FLEX+DMFT, spontaneous $d_{x^2-y^2}\pm i d_{xy}$7 Pomeranchuk order enhances $d_{x^2-y^2}\pm i d_{xy}$8 and yields a real $d_{x^2-y^2}\pm i d_{xy}$9 gap, with no evidence for a complex $d+id$0 state in the parameter range studied (Sayyad et al., 2021). By contrast, loop-current-induced pair hopping in kagome metals stabilizes a chiral d-wave state, while coexistence with charge order admixes an $d+id$1-wave component and makes the state nematic-chiral rather than purely chiral (Tazai et al., 6 Aug 2025).

The isotropic triangular-lattice Hubbard model provides a quantitative benchmark for this competition. Using VCA at $d+id$2 and half filling, the ground state for $d+id$3 is nematic $d+id$4, while the chiral $d+id$5 states are consistently higher in energy. Near $d+id$6, the condensation energy relative to the normal paramagnet is $d+id$7–$d+id$8, whereas the chiral–nematic splitting is $d+id$9–$d_{xy}$0; above the Mott point, no d-wave superconducting solution is realized (Yamada, 2024). This is a concrete example in which symmetry permits chirality but microscopic energetics select nematicity.

4. Topology, electromagnetic response, and vortex textures

Chiral d-wave states are topological because the complex order parameter endows Bogoliubov quasiparticles with nontrivial Berry curvature. In SrPtAs, the chiral $d_{xy}$1 state yields fully gapped bands with $d_{xy}$2 and a nodal band with $d_{xy}$3 for $d_{xy}$4 and $d_{xy}$5 outside that interval; the same state hosts Majorana-Weyl nodes in the bulk, chiral Majorana surface bands, Fermi arcs on a $d_{xy}$6 surface, and a thermal Hall response

$d_{xy}$7

set by the $d_{xy}$8-resolved Chern numbers (Fischer et al., 2013). On the honeycomb lattice near the Mott state, the chiral $d_{xy}$9 state carries Chern number $d+id$0 and guarantees two co-propagating chiral edge modes (Black-Schaffer et al., 2014).

Moiré chiral d-wave superconductivity adds further structure. In twisted bilayer graphene, the fully gapped chiral state has BdG Chern number $d+id$1 per spin block and total Chern number $d+id$2 including spin degeneracy; the same state supports spontaneous bulk circulating supercurrent, a magnetic dipole $d+id$3 per moiré cell, and a toroidal dipole $d+id$4 for a representative pairing scale (Wu, 2018). These are not generic features of all chiral d-wave states; they arise here from moiré band geometry and layer counterflow.

The topology of chiral d-wave order also manifests in vortex matter. For a chiral $d+id$5-wave state with $d+id$6, the phase-winding constraint relates the global vorticity $d+id$7 of the dominant component and the local winding $d+id$8 of the induced subdominant component through $d+id$9, so a coreless vortex requires $$\Delta_{d1}(\mathbf{k}) \propto k_x^2-k_y^2,\qquad \Delta_{d2}(\mathbf{k}) \propto 2k_xk_y,$$0 and therefore $$\Delta_{d1}(\mathbf{k}) \propto k_x^2-k_y^2,\qquad \Delta_{d2}(\mathbf{k}) \propto 2k_xk_y,$$1. The resulting quadruply quantized coreless vortices exhibit ring-shaped subgap LDOS peaks, paramagnetic ring currents, and a strong asymmetry between positive and negative magnetic fields, which directly expose time-reversal breaking and the Chern number (Holmvall et al., 2022). A nematic d-wave state, by contrast, is topologically trivial in this sense and does not produce the same field-direction asymmetry.

Topology does not imply isotropy. In the doped triangular $$\Delta_{d1}(\mathbf{k}) \propto k_x^2-k_y^2,\qquad \Delta_{d2}(\mathbf{k}) \propto 2k_xk_y,$$2–$$\Delta_{d1}(\mathbf{k}) \propto k_x^2-k_y^2,\qquad \Delta_{d2}(\mathbf{k}) \propto 2k_xk_y,$$3–$$\Delta_{d1}(\mathbf{k}) \propto k_x^2-k_y^2,\qquad \Delta_{d2}(\mathbf{k}) \propto 2k_xk_y,$$4 model, DMRG identifies an isotropic chiral topological superconductor with $$\Delta_{d1}(\mathbf{k}) \propto k_x^2-k_y^2,\qquad \Delta_{d2}(\mathbf{k}) \propto 2k_xk_y,$$5, a nematic chiral topological superconductor that still has $$\Delta_{d1}(\mathbf{k}) \propto k_x^2-k_y^2,\qquad \Delta_{d2}(\mathbf{k}) \propto 2k_xk_y,$$6, a critical chiral phase with $$\Delta_{d1}(\mathbf{k}) \propto k_x^2-k_y^2,\qquad \Delta_{d2}(\mathbf{k}) \propto 2k_xk_y,$$7, and a time-reversal-symmetric nematic d-wave phase with $$\Delta_{d1}(\mathbf{k}) \propto k_x^2-k_y^2,\qquad \Delta_{d2}(\mathbf{k}) \propto 2k_xk_y,$$8 (Huang et al., 2021). This establishes that nematicity and chirality are not mutually exclusive: rotational symmetry breaking can coexist with nonzero Chern number.

5. Model systems and material realizations

Microscopic studies do not point to a single universal outcome. Instead, the same $$\Delta_{d1}(\mathbf{k}) \propto k_x^2-k_y^2,\qquad \Delta_{d2}(\mathbf{k}) \propto 2k_xk_y,$$9 manifold is selected differently by frustration, multiband fermiology, flat-band physics, spin chirality, loop-current order, or external fields.

System	Dominant state reported	Characteristic result
Isotropic triangular-lattice Hubbard (Yamada, 2024)	Nematic $d_{xy}$00 ground state	Chiral $d_{xy}$01 is quasi-stable and higher by $d_{xy}$02–$d_{xy}$03
SrPtAs (Fischer et al., 2013)	Chiral $d_{xy}$04 $d_{xy}$05	TRS breaking, Majorana-Weyl nodes, thermal Hall response
Honeycomb lattice near the Mott state (Black-Schaffer et al., 2014)	Chiral $d_{xy}$06	Fully gapped away from half filling, Chern number $d_{xy}$07
TBG moiré superconductivity (Wu, 2018)	Chiral topological d-wave	Total Chern number $d_{xy}$08, spontaneous vortex–antivortex COMM texture
Multiple-flat-band TBG limit (Liu et al., 2024)	Nodal nematic Euler d-wave	Inter-eigen-band pairing favors nematicity over chiral order
Partially flat triangular lattice (Sayyad et al., 2021)	Real $d_{xy}$09–$d_{xy}$10–$d_{xy}$11 nematic mixture	Nematicity enhances $d_{xy}$12; no chiral solution found
Doped triangular $d_{xy}$13–$d_{xy}$14–$d_{xy}$15 model (Huang et al., 2021)	Isotropic chiral, nematic chiral, and nematic d-wave phases	Chern numbers $d_{xy}$16, $d_{xy}$17, and $d_{xy}$18, respectively
Kagome AV$d_{xy}$19Sb$d_{xy}$20 (Tazai et al., 6 Aug 2025)	Nematic chiral d-wave	Loop-current order drives chirality; disorder can switch to $d_{xy}$21 wave

The triangular-lattice Hubbard result is closely tied to organic salts. Using $d_{xy}$22 eV, the calculated zero-temperature energy gain of the $d_{xy}$23 state over the normal state, $d_{xy}$24–$d_{xy}$25, corresponds to $d_{xy}$26–$d_{xy}$27 K, while the experimental superconducting transition in $d_{xy}$28-(BEDT-TTF)$d_{xy}$29Cu$d_{xy}$30(CN)$d_{xy}$31 is $d_{xy}$32 K; the paper regards this as semi-quantitatively consistent because the transition temperature is determined by finite-temperature free energies rather than the zero-temperature condensation energy alone (Yamada, 2024). The same work further argues that the predicted $d_{xy}$33 node orientation agrees with thermal conductivity tensor analysis and STM.

SrPtAs represents the contrasting case in which experiments already indicate broken time-reversal symmetry at the onset of superconductivity, and the multiband fRG analysis identifies a chiral $d_{xy}$34 instability driven mainly by the K/K$d_{xy}$35 pockets near a van Hove singularity, with the outer band contributing about $d_{xy}$36 of the density of states at the Fermi level (Fischer et al., 2013). The honeycomb near-Mott analysis reaches a similar chiral conclusion, but by strong-coupling arguments: RMFT and DQMC both favor $d_{xy}$37 over nematic d-wave, and attempts to construct mixed-chirality or real-space zero-chirality supercells up to eight sites are energetically unfavorable (Black-Schaffer et al., 2014).

Twisted bilayer graphene has become the principal setting in which the distinction between nematic and chiral d-wave is genuinely unsettled. One microscopic route yields a fully gapped chiral topological state with spontaneous currents and moiré-center-of-mass vortices (Wu, 2018). Another, developed in the multiple-flat-band limit, finds that inter-eigen-band pairing stabilizes a nodal nematic Euler state over the chiral alternative (Liu et al., 2024). A third result shows that an in-plane magnetic field can hybridize the two zero-field chiralities and induce a transport response with twofold anisotropy, so field-induced nematicity need not imply a nematic zero-field ground state (Yu et al., 2021).

6. Experimental discrimination and open issues

The most direct separation between nematic and chiral d-wave order is still symmetry-resolved spectroscopy and transport. A nematic state is expected to show line or point nodes, twofold anisotropy, and no spontaneous time-reversal-symmetry breaking, whereas a chiral state is expected to show time-reversal breaking and, in many settings, a full gap or only topologically protected point nodes. For the triangular organic material, the proposed distinction is explicit: STM or QPI should reveal two-mirror-symmetric nodal patterns for $d_{xy}$38, while a fully gapped spectrum would instead indicate chiral $d_{xy}$39; muSR, polar Kerr effect, and zero-field Hall probes were proposed as tests of time-reversal breaking, and strain or pressure tuning near $d_{xy}$40–$d_{xy}$41 was proposed as a way to expose the small chiral–nematic splitting (Yamada, 2024).

Time-reversal-sensitive probes have already been decisive in some materials. In SrPtAs, $d_{xy}$42SR detects time-reversal-symmetry breaking at $d_{xy}$43, which strongly disfavors a real nematic $d_{xy}$44 state and supports the chiral interpretation (Fischer et al., 2013). In kagome AV$d_{xy}$45Sb$d_{xy}$46, the proposed loop-current mechanism naturally connects chiral quasiparticle interference, nonreciprocal transport, giant thermal Hall signals, and a superconducting diode effect to a nematic chiral d-wave condensate; the same paper predicts that a small amount of impurities can suppress the unconventional chiral channel and reveal an isotropic $d_{xy}$47-wave state instead (Tazai et al., 6 Aug 2025).

A central interpretive issue is that twofold anisotropy does not by itself determine the zero-field order parameter. In magic-angle twisted bilayer graphene, the in-plane-field theory predicts that orbital coupling hybridizes the two chiral superconducting components above a field scale

$d_{xy}$48

and the paraconductivity becomes

$d_{xy}$49

This implies that the observed $d_{xy}$50 transport anisotropy near $d_{xy}$51 does not rule out a chiral ground state; it may instead reflect field-induced nematicity of an underlying chiral superconductor (Yu et al., 2021).

Several open problems remain model-specific rather than conceptual. In the triangular-lattice Hubbard study, the chiral state is described as topologically equivalent across two real-space constructions, but explicit topological invariants were not computed (Yamada, 2024). The multiple-flat-band TBG mechanism is developed in the chiral limit of a heavy-fermion model, and the authors note that real TBG departs from that limit and includes Coulomb effects beyond the separable phonon-mediated interaction (Liu et al., 2024). The triangular $d_{xy}$52–$d_{xy}$53–$d_{xy}$54 phase diagram is based on finite-width cylinders, and the paper explicitly notes that finite-size geometry can affect gap estimates and shift higher-order transition indicators even though the Chern assignments are robust under flux insertion (Huang et al., 2021).

Taken together, these studies indicate that the decisive question is not whether the $d_{xy}$55 manifold permits both nematic and chiral d-wave order—it does—but which perturbations lift the near-degeneracy in a given material. Frustration, flat-band multiband pairing, density-wave or gauge-field fluctuations, spin chirality, loop-current order, strain, magnetic field, and disorder all act as selectors. Nematic chiral d-wave superconductivity is therefore best understood not as a single phase with a universal phenomenology, but as a symmetry-organized landscape in which time-reversal breaking, rotational symmetry breaking, topology, and multicomponent mixing can either compete or coexist.