---
title: Direction-Selective Nematic Criticality
url: https://www.emergentmind.com/topics/direction-selective-nematic-criticality
type: topic
---

# Direction-Selective Nematic Criticality

Searching arXiv for recent and foundational papers on direction-selective nematic criticality and related nemato-elasticity.
arxiv_search.query({"search_query":"all:\"direction-selective nematic criticality\" OR all:\"nematic criticality\" nemato-elastic compatibility", "start": 0, "max_results": 10, "sort_by": "submittedDate", "sort_order": "descending"})
Searching arXiv specifically for elastic compatibility constrained nematic criticality and strain-tuned nematic QCPs.
arxiv_search.query({"search_query":"all:\"elastic compatibility\" nematic criticality OR cat:cond-mat.str-el AND all:\"strain-tuned quantum criticality\" nematic", "start": 0, "max_results": 10, "sort_by": "relevance", "sort_order": "descending"})
Direction-selective nematic criticality is the regime in which critical nematic fluctuations do not soften uniformly in momentum space or across symmetry channels, but instead remain critical only for symmetry-selected directions, components, or sectors of the nematic order parameter. In electronic systems, nematic order denotes spontaneous breaking of rotational point-group symmetry without primarily breaking translational symmetry; depending on lattice and microscopic realization, the order parameter may be Ising-like, Potts-like, or a higher-rank traceless tensor. The modern formulation of the subject emphasizes that this selectivity is not merely a materials-specific anisotropy effect. In metals and crystals, it can arise from the symmetry of the nematic form factor, from Fermi-surface geometry, from bilinear coupling to strain, and, more fundamentally, from Saint Venant compatibility constraints that reorganize the nematic fluctuation spectrum into soft compatible modes and suppressed incompatible modes [1610.06168][2507.23754].

## 1. Symmetry content and general definition

Direction selectivity enters nematic criticality because the nematic instability is itself orientational. An electronic nematic phase selects one spatial direction over another while preserving translation symmetry. In tetragonal iron-based superconductors this often appears as an Ising nematic order with either \(B_{1g}\) or \(B_{2g}\) symmetry; in the strain-tuned Ba(Fe\(_{1-x}\)Co\(_x\))\(_2\)As\(_2\) study, the nematic order is described as Ising nematic with \(B_{2g}\) symmetry, and the electronic system “chooses” one in-plane crystalline direction over the orthogonal one [2003.12202]. In optimally doped Fe-based superconductors, the critical response is not isotropic across symmetry channels: the large, often diverging elastoresistive response appears in the \(B_{2g}\) channel, whereas a representative \(B_{1g}\) measurement does not show a comparable divergence [1503.00402].

In a more general tensor formulation, the nematic order parameter is a real, symmetric, traceless rank-two object. In three dimensions it can be represented by five components, as in the \(\ell=2\) quadrupolar basis used for nemato-elastic theories and for quadratic-band-touching semimetals. In that setting, direction selectivity means that only a subset of the five-component manifold can become critical, and only along particular momentum directions or with particular director orientations [2507.23754][1503.04242].

The expression “direction-selective” therefore has more than one technically distinct meaning. In some experiments it refers to symmetry-channel selectivity, such as \(B_{2g}\) versus \(B_{1g}\) response in tetragonal systems [1503.00402]. In metal-plus-lattice theories it refers to momentum-space selectivity, where only specific high-symmetry wavevector directions remain critical after coupling to strain [1610.06168]. In compatibility-based formulations it refers to a sharper statement: the critical manifold is a direction-dependent compatible subspace of the full nematic tensor field, while the remaining orthogonal components are noncritical or gapped [2507.23754].

## 2. Electron-only and fluctuation-driven formulations

An important starting point is the electron-only description of nematic quantum criticality. For a two-dimensional interacting Fermi system with forward scattering in the \(d\)-wave channel,
\[
H = \sum_{k,\sigma} \epsilon_k\, c^\dagger_{k\sigma} c_{k\sigma} + \frac{1}{2L}\sum_{k,k',q} f_{kk'}(q)\, n_k(q)\, n_{k'}(-q),
\]
with
\[
f_{kk'}(q) = -g(q)\, d_k d_{k'}, \qquad d_k=\cos k_x-\cos k_y,
\]
the order parameter
\[
\phi = \frac{g}{L}\sum_k d_k\, n_k
\]
breaks rotational symmetry but preserves translation symmetry [1101.1221]. Mean-field theory predicts a nematic dome near van Hove filling, but the Landau expansion is singular there and becomes unreliable because the coefficients are negative at all orders at low temperature [1101.1221].

The non-perturbative functional renormalization-group treatment of this model shows that order-parameter fluctuations can shrink and even eliminate the mean-field nematic dome. For a critical cutoff \(\Lambda_0^*\), the entire \((\mu,T)\) phase diagram is disordered while an isolated quantum critical point remains at
\[
(\mu,T) = (\mu_{\rm qc}^*,0),
\]
with
\[
\mu_{\rm qc}^* \approx -1.404,\qquad \Lambda_0^* \approx e^{-0.2574}.
\]
At that isolated point the system exhibits quantum critical behavior without any intervening ordered nematic regime, and the zero-temperature correlation-length exponent becomes \(\nu=1\), not the naive Gaussian \(\nu=\tfrac12\) [1101.1221]. This establishes that direction-selective nematic criticality need not imply a finite ordered dome; fluctuations can leave only the critical singularity.

A distinct electron-only setting is the three-dimensional quadratic-band-touching semimetal. There the low-energy fermions are four-component spinors coupled to a five-component tensor nematic field through a Gross-Neveu-Yukawa action,
\[
L_{\psi\phi}=g\,\phi_a\,\psi^\dagger\gamma_a\psi,
\]
with bosonic cubic invariant \(\lambda\,\mathrm{Tr}\,T^3\) [1503.04242]. Near \(d=4\), the theory admits a stable interacting fixed point describing a continuous transition from the semimetal to a nematic Mott insulator. The ordered phase breaks rotational symmetry but preserves time-reversal symmetry and opens an anisotropic gap, making the critical point a quantum analogue of the classical isotropic-to-nematic transition in liquid crystals [1503.04242].

These electron-centric theories already show two central facts. First, nematic criticality is highly sensitive to fluctuation effects and to the tensor structure of the order parameter. Second, direction selectivity is not reducible to a simple low-order Ising picture: it can appear in two-dimensional \(d\)-wave metals, in five-component three-dimensional tensor theories, and even in regimes without stable long-range nematic order [1101.1221][1503.04242].

## 3. Nemato-elasticity, lattice coupling, and compatibility constraints

A major development in the subject is the recognition that electronic nematic criticality in a crystal is inseparable from elasticity. The symmetry-allowed bilinear coupling between the electronic nematic field and strain modifies the critical theory qualitatively. In the orthorhombic case considered for metals near an Ising-nematic quantum critical point,
\[
\mathcal{H}_{\rm nem-latt}=\lambda \int d\mathbf{r}\,\phi(\mathbf{r})\,\varepsilon(\mathbf{r}),
\]
with
\[
\varepsilon(\mathbf{r})=\varepsilon_{xx}(\mathbf{r})-\varepsilon_{yy}(\mathbf{r}),
\]
the lattice shifts the critical point from \(r=0\) to
\[
r_0=\lambda^2\nu_0/C_0,
\]
and renders the nematic mass angle dependent [1610.06168]. At the shifted critical point, the renormalized mass vanishes only for the two high-symmetry directions
\[
\hat q_{1,2}\equiv \frac{(\hat q_x\pm \hat q_y)}{\sqrt 2},
\]
while for all other directions it remains finite. The resulting susceptibility near a critical direction is strongly anisotropic,
\[
\chi^{-1}(\mathbf{q}\approx \hat q_1)=\nu_0^{-1}\left[r_0\frac{q_2^2+q_z^2}{q_1^2}+\frac{q_1^2}{k_F^2}\right],
\]
so the critical scaling is itself direction dependent [1610.06168].

The compatibility-based formulation developed later sharpens this picture and generalizes it beyond a particular lattice symmetry. In linear elasticity, strain must derive from a displacement field,
\[
\varepsilon_{ij}(\mathbf{x})=\frac12\left[\partial_i u_j(\mathbf{x})+\partial_j u_i(\mathbf{x})\right],
\]
and therefore satisfies the Saint Venant condition
\[
\mathrm{inc}(\varepsilon)_{ij}\equiv \epsilon_{ikl}\epsilon_{jmn}\partial_k\partial_m\varepsilon_{ln}=0.
\]
Because the electronic nematic tensor \(\varphi_{ij}\) couples bilinearly to strain, it inherits this compatibility structure [2507.23754]. The technically decisive step is the introduction of a momentum-dependent co-rotating helical basis. In that basis, the five helical nematic amplitudes \(\Phi_a\) split into a compatible critical doublet and suppressed incompatible modes. After integrating out compatible elastic degrees of freedom, the quadratic effective theory becomes
\[
\mathcal F_{\rm eff}^{(2)}= \frac{1}{2V}\sum_{\mathbf q}\Big[ \tilde r_c(\mathbf q)\sum_{\alpha=2,3}|\Phi_\alpha|^2 +\tilde r_1(\mathbf q)|\Phi_1|^2 +r(\mathbf q)\sum_{\alpha=4,5}|\Phi_\alpha|^2 \Big],
\]
with
\[
\tilde r_c(\mathbf q)=r(\mathbf q)-\frac{\lambda_0^2}{\mu}, \qquad \tilde r_1(\mathbf q)=r(\mathbf q)-\frac{\lambda_0^2\varrho}{\mu}.
\]
Only \(\Phi_2\) and \(\Phi_3\) soften fully; \(\Phi_1\) is mass enhanced and \(\Phi_4,\Phi_5\) are gapped or suppressed [2507.23754].

This decomposition provides the strongest current statement of direction-selective nematic criticality. The selectivity is not attributed to accidental phonon anisotropy, but to geometric integrability constraints of strain. In the isotropic-medium formulation, the critical manifold is reduced from the five-dimensional \(\ell=2\) space to a two-dimensional compatible subspace with effective \(\mathrm{SO}(2)\) symmetry [2507.23753]. In the orbital basis, a given nematic component such as \(\varphi_{2xy}\) is critical only for those momentum directions where its overlap with the soft helical modes is complete; for other directions it overlaps partly or wholly with the suppressed \(\Phi_1,\Phi_4,\Phi_5\) sectors and becomes massive [2507.23754].

A further implication concerns defects. When defects are present, the gauge-invariant strain is
\[
\mathscr E_{ij}(\mathbf x)=\varepsilon_{ij}(\mathbf x)-\varepsilon^p_{ij}(\mathbf x),
\]
and the defect-generated random conjugate field acts primarily on the noncritical helical sectors rather than on the critical compatible doublet [2507.23754]. This suggests why thermodynamic criticality can remain mean-field-like while local probes observe strong domain formation and pinning effects: the critical and defect-sensitive sectors are not the same [2507.23753].

## 4. Strain tuning and symmetry-selective experiments in Fe-based superconductors

Iron-based superconductors provide some of the clearest experimental realizations of nematic quantum criticality. A central strain-tuned study was performed on Ba(Fe\(_{1-x}\)Co\(_x\))\(_2\)As\(_2\), where uniaxial stress along \([100]\) generates a nearly continuously tunable \(\epsilon_{xx}\) and shifts the coupled nematic/structural transition temperature \(T_S\) according to
\[
T_S(\epsilon_{xx}) = T_0 + \alpha \epsilon_{xx} + \beta \epsilon_{xx}^2.
\]
Here the linear coefficient is dominated by the \(A_{1g}\) contribution, while the quadratic coefficient is associated with the \(B_{1g}\) response; hydrostatic-pressure measurements show that the quadratic \(A_{1g}\) coefficient \(a'\) is negligible, allowing the two channels to be disentangled [2003.12202].

The quantum-critical scaling ansatz
\[
T_S \sim (g_C-g)^\theta
\]
implies
\[
\frac{\alpha}{T_0} \sim -T_0^{-1/\theta}, \qquad \frac{\beta}{T_0} \sim -T_0^{-1/\theta}
\]
as \(T_0\to 0\). Experimentally, both \(\alpha/T_0\) and \(\beta/T_0\) increase strongly and appear to diverge as cobalt concentration approaches \(x_c\sim 6.7\%\). Log-log fits give
\[
1/\theta = 1.75 \pm 0.2 \quad (A_{1g}), \qquad 1/\theta = 1.87 \pm 0.7 \quad (B_{1g}),
\]
with average
\[
\theta = 0.56 \pm 0.07,
\]
while a scaling-collapse analysis yields
\[
\theta = 0.52 \pm 0.02.
\]
The phase boundary also collapses in the form
\[
T_S^{1/\theta} - T_0^{1/\theta} = a\,\epsilon_{xx} + b\,\epsilon_{xx}^2,
\]
for compositions \(x\approx 4.8\%\) to \(6.2\%\), with transition temperatures from about \(25\) K to \(70\) K [2003.12202]. In this setting, direction selectivity appears through symmetry decomposition of strain response and through the observation that the same exponent governs inequivalent symmetry channels.

A complementary experimental route uses differential elastoresistance. In five optimally doped Fe-based superconductors, the \(B_{2g}\) elastoresistivity coefficient \(2m_{66}\), which serves as a proxy for the nematic susceptibility, grows strongly on cooling and often follows a Curie-Weiss form,
\[
2m_{66} = 2m_{66}^0 + \frac{\lambda}{a(T-T^*)}.
\]
For optimally doped BaFe\(_2\)(As\(_{0.68}\)P\(_{0.32}\))\(_2\), a log-log fit gives
\[
|2m_{66}-2m_{66}^0| \sim (T-T^*)^{-\gamma}, \qquad \gamma = 0.985 \pm 0.005,
\]
with \(T^*\) close to zero, consistent with a nematic quantum critical point [1503.00402]. The direction-selective aspect here is explicitly symmetry selective: the pronounced divergence is observed in the \(B_{2g}\) channel, while a representative \(B_{1g}\) check does not show comparable criticality [1503.00402].

| System | Tuning parameter | Reported critical signature |
|---|---|---|
| Ba(Fe\(_{1-x}\)Co\(_x\))\(_2\)As\(_2\) | cobalt concentration and uniaxial strain \(\epsilon_{xx}\) | divergence of \(\alpha/T_0\) and \(\beta/T_0\); scaling collapse of \(T_S^{1/\theta}\) [2003.12202] |
| Optimally doped Fe-based superconductors | temperature under anisotropic strain | large \(B_{2g}\) elastoresistivity \(2m_{66}\), often Curie-Weiss-like [1503.00402] |
| FeSe\(_{1-x}\)S\(_x\) | sulfur substitution and pressure | two distinct linear-\(T\) resistivity components linked separately to nematic and magnetic criticality [2106.08821] |

FeSe\(_{1-x}\)S\(_x\) adds a distinct lesson. Near the nematic quantum critical point at \(x_c\simeq 0.17\), the low-temperature resistivity contains a large linear-\(T\) component associated with nematic critical fluctuations, while pressure induces a smaller higher-temperature linear-\(T\) component that grows on approaching magnetism. The coefficients evolve oppositely with pressure, and the study interprets this as evidence that nematic and magnetic critical fluctuations are completely decoupled [2106.08821]. This does not by itself establish direction selectivity in momentum space, but it sharply distinguishes nematic criticality from magnetic criticality in a material where the two are often conflated.

## 5. Momentum-space selectivity, Fermi-surface geometry, and thermodynamics

When the lattice is included explicitly in metallic theories, direction-selective criticality has direct consequences for quasiparticles and thermodynamics. The angle-dependent mass generated by nemato-elastic coupling implies that only fluctuations near special directions \(\hat q_{1,2}\) remain critical [1610.06168]. Whether those modes are Landau damped then depends on Fermi-surface geometry and on the nematic form factor \(h_{\mathbf k}\sim \cos k_x-\cos k_y\).

For cuprate-like Fermi surfaces, the critical directions intersect the Fermi surface where the form factor vanishes, so the damping of the only critical bosons is suppressed and the dynamics becomes ballistic,
\[
D(\hat q_1,i\Omega_n)\propto \frac{\Omega_n^2}{(v_F q_1)^2},
\]
with \(z=2\). In that case the entire Fermi surface remains cold [1610.06168]. For Fe-based superconductors, the electron pockets near \((\pi,0)\) and \((0,\pi)\) have nonvanishing form factor at the relevant tangency points, so the critical boson remains Landau damped,
\[
D(\hat q_1,i\Omega_n)\propto \frac{|\Omega_n|}{v_F q_1},
\]
and only hot spots on the electron pockets survive, while the hole pockets remain cold [1610.06168].

The low-temperature thermodynamics are correspondingly less singular than in electron-only Hertz-Millis treatments. The fluctuation free energy is
\[
F=\frac{T}{2}\sum_{\mathbf{q},\Omega_n}\log \chi^{-1}(\mathbf{q},i\Omega_n),
\]
but the broad noncritical sector dominates phase space. Below
\[
T_{\rm FL}\sim r_0^{3/2}E_F,
\]
the specific-heat coefficient remains Fermi-liquid-like,
\[
\gamma(T)\propto \frac{1}{r_0^{1/2}},
\]
with critical corrections subleading [1610.06168]. This is a major correction to the common electron-only expectation that a nematic quantum critical point in a metal generically makes nearly the whole Fermi surface hot and produces non-Fermi-liquid thermodynamics.

The compatibility-based tensor theory reaches a related conclusion by a different route. The suppression of incompatible modes increases the effective dimensional rigidity of the critical theory, while local orbital observables mix long-ranged critical helical modes with short-ranged massive ones through
\[
\chi_{aa}(\mathbf q) = \frac{\hat Q_{2,a}^2(\hat q)+\hat Q_{3,a}^2(\hat q)}{r(\mathbf q)-\lambda_0^2/\mu} + \frac{\hat Q_{1,a}^2(\hat q)}{r(\mathbf q)-\lambda_0^2\varrho/\mu} + \frac{\hat Q_{4,a}^2(\hat q)+\hat Q_{5,a}^2(\hat q)}{r(\mathbf q)}.
\]
A local measurement in the orbital basis therefore does not isolate a single critical mode; it probes a nonlocal superposition of compatible critical and incompatible massive sectors [2507.23754]. This suggests a natural explanation for why local probes may detect inhomogeneity or apparently incomplete softening even when a sharp critical manifold exists.

## 6. Broader extensions, multicritical variants, and unresolved issues

Direction-selective nematic criticality is not limited to Ising-like tetragonal metals. In electronic three-state Potts-nematic systems with threefold rotational symmetry, the zero-strain quantum transition is first-order because of the cubic invariant. Uniaxial strain acts as a conjugate field,
\[
-\gamma \varepsilon\, \phi \cos(2\theta-2\alpha),
\]
and generates different zero-temperature critical structures depending on the sign of strain [2303.00685]. For \(\tilde\varepsilon<0\), the system exhibits a symmetry-preserving meta-nematic transition terminating at a quantum critical end-point,
\[
\tilde\varepsilon_{\text{QCEP}}=-\frac{1}{27}, \qquad a_{\text{QCEP}}=\frac{1}{3}.
\]
For \(\tilde\varepsilon>0\), the director unlocks, spontaneously breaks an in-plane twofold symmetry, and the first-order line changes to second-order at a quantum tricritical point,
\[
a_{\text{QTCP}}=\frac{7}{16},\qquad \tilde\varepsilon_{\text{QTCP}}=\frac{3}{16}.
\]
Once coupled to itinerant electrons, both strain-tuned Potts critical points acquire overdamped dynamics similar to the Ising-nematic case, with
\[
\delta\Pi(q)\sim -\frac{|\Omega|}{|\mathbf q|},
\]
and \(z=3\) except at cold spots where \(z=2\) [2303.00685]. This indicates that strain can reduce a three-state nematic manifold to an effectively one-component critical sector, but in a way that preserves the asymmetry between compressive and tensile tuning.

Another benchmark system is Sr\(_3\)Ru\(_2\)O\(_7\), where thermal-expansion measurements near the field-tuned metamagnetic regime reveal a second-order nematic transition for \(B\parallel c\), with critical behavior compatible with the two-dimensional Ising universality class, although not conclusively so [1212.0948]. A slight field tilt produces direct thermodynamic evidence of broken in-plane fourfold symmetry inside the nematic region and a weak residual anisotropy above the upper first-order boundary \(B_{c2}\), suggesting that the symmetry breaking extends somewhat beyond the nominal dome [1212.0948]. This system shows that direction-selective nematicity can develop in close relation to a metamagnetic quantum critical point, even when the microscopic setting differs markedly from Fe-based superconductors.

Several recurrent misconceptions are corrected by the current literature. One is that nematic criticality in metals should generically be isotropic and produce non-Fermi-liquid behavior over nearly the entire Fermi surface; lattice coupling and compatibility constraints show otherwise [1610.06168][2507.23754]. Another is that strong nematic criticality necessarily implies an adjacent ordered nematic region; the fluctuation-driven two-dimensional FRG example demonstrates a quantum critical point without order [1101.1221]. A third is that domain formation and mean-field-like thermodynamic criticality are contradictory; compatibility-based nemato-elastic theory explains their coexistence by separating compatible critical modes from incompatible defect-coupled modes [2507.23753].

A plausible implication is that “direction-selective nematic criticality” should now be understood less as a narrow label for one material class than as a unifying principle across several settings: strain-tuned Fe-based superconductors, lattice-coupled metallic nematicity, compatibility-constrained tensor theories, Potts-nematic systems, quadratic-band-touching semimetals, and metamagnetic ruthenates all exhibit some form of critical-sector reduction. What differs from case to case is the mechanism of the reduction—symmetry-channel selection, Fermi-surface filtering, strain-induced component selection, or geometric compatibility—but the common outcome is the same: only a restricted subset of the nominal nematic degrees of freedom becomes truly critical [2003.12202][1503.00402][2507.23754].

Source: https://www.emergentmind.com/topics/direction-selective-nematic-criticality