Charge and Spin Density Waves

A density wave is a state in which a charge, spin, orbital, or bond observable develops a periodic component at a nonzero ordering wavevector $\mathbf Q$. A charge-density wave (CDW) modulates a time-reversal-even electronic observable; a spin-density wave (SDW) modulates magnetization and is odd under time reversal. Both break translations when $\mathbf Q$ is not a reciprocal-lattice vector, reconstruct electronic states by mixing momenta separated by $\mathbf Q$, and support collective fluctuations of their amplitude and phase.

Density waves are identified by their operator, wavevector, symmetry, correlation length, and dynamics. A nested Fermi-surface sketch, a partial gap, or a resistivity anomaly is not by itself a density-wave diagnosis.

Canonical Scope

Order Parameters owns the general construction of symmetry-breaking fields. Long-Range Order, Structure Factors, and Equal-Time Correlations own thermodynamic-limit and finite-size tests. Itinerant Magnetism owns the detailed Stoner and matrix-susceptibility route to material SDWs, including chromium. BCS Theory owns uniform weak-coupling superconductivity.

This page is the canonical home for comparing charge and spin density waves in quantum materials: their symmetry and harmonics, nesting and Peierls limits, band reconstruction, collective modes, experimental evidence, and competition or coexistence with superconductivity.

Periodic Charge and Spin Modulation

Write the microscopic density as a lattice-periodic background plus slowly varying Fourier components:

$$\begin{aligned} \langle n(\mathbf r)\rangle ={}& n_{\mathrm{per}}(\mathbf r) + \operatorname{Re} \left[ \rho_{\mathbf Q}(\mathbf r) e^{i\mathbf Q\cdot\mathbf r} \right] +\cdots , \\ \langle\mathbf S(\mathbf r)\rangle ={}& \mathbf M_{\mathrm{per}}(\mathbf r) + \operatorname{Re} \left[ \mathbf M_{\mathbf Q}(\mathbf r) e^{i\mathbf Q\cdot\mathbf r} \right] +\cdots . \end{aligned}$$

Reality requires $\rho_{-\mathbf Q}=\rho_{\mathbf Q}^{*}$ and $\mathbf M_{-\mathbf Q}=\mathbf M_{\mathbf Q}^{*}$. The order parameters may carry orbital, sublattice, or bond indices that are suppressed here. Those internal form factors can be decisive: a bond-density wave may strongly modulate hopping or kinetic energy while producing only a small onsite charge modulation.

Under a lattice translation by $\mathbf a$,

$$\rho_{\mathbf Q} \mapsto e^{i\mathbf Q\cdot\mathbf a}\rho_{\mathbf Q}, \qquad \mathbf M_{\mathbf Q} \mapsto e^{i\mathbf Q\cdot\mathbf a}\mathbf M_{\mathbf Q}.$$

Time reversal leaves the physical charge profile unchanged and reverses the spin profile. For the fixed-$\mathbf Q$ coefficients in the real-space expansion above,

$$\mathcal T: \qquad \rho_{\mathbf Q}\mapsto\rho_{\mathbf Q}, \qquad \mathbf M_{\mathbf Q}\mapsto-\mathbf M_{\mathbf Q}.$$

At the operator level, antiunitarity instead gives $\mathcal T\widehat\rho_{\mathbf Q}\mathcal T^{-1}=\widehat\rho_{-\mathbf Q}$ and $\mathcal T\widehat{\mathbf S}_{\mathbf Q}\mathcal T^{-1}=-\widehat{\mathbf S}_{-\mathbf Q}$. Taking the transformed expectation value supplies a second complex conjugation, producing the fixed-$\mathbf Q$ rule above. Spin rotations act on $\mathbf M_{\mathbf Q}$ but not on a scalar CDW. Thus a collinear SDW generally breaks spin rotation, time reversal, and translation, whereas a nonmagnetic scalar CDW need only break translation and perhaps point-group symmetries.

Longitudinal, spiral, and stripe order

A real vector amplitude gives a collinear SDW,

$$\langle\mathbf S(\mathbf r)\rangle = \mathbf M_0\cos(\mathbf Q\cdot\mathbf r+\phi).$$

If the real and imaginary parts of $\mathbf M_{\mathbf Q}$ are orthogonal, the spin direction rotates in space and the state is spiral or helical. A stripe usually denotes unidirectional charge or spin order, often with an accompanying breaking of rotational symmetry. In a doped antiferromagnet, charge-rich domain walls can separate antiphase spin regions; this is a strong-coupling organization and need not originate from weak-coupling Fermi-surface nesting.

Harmonics and coupled order

Nonlinear couplings generate harmonics. For a collinear SDW,

$$\lvert\mathbf M_0\cos(\mathbf Q\cdot\mathbf r+\phi)\rvert^2 = \frac{\lvert\mathbf M_0\rvert^2}{2} \left[ 1+\cos(2\mathbf Q\cdot\mathbf r+2\phi) \right],$$

so a charge or lattice modulation at $2\mathbf Q$ is symmetry-allowed. A Landau coupling that encodes this relation is

$$F_{\mathrm{coup}} = \lambda\, \rho_{2\mathbf Q}^{*} \left( \mathbf M_{\mathbf Q}\cdot\mathbf M_{\mathbf Q} \right) +\mathrm{c.c.}$$

For a circular spiral, $\mathbf M_{\mathbf Q}\cdot\mathbf M_{\mathbf Q}=0$, so the induced $2\mathbf Q$ scalar harmonic can vanish even though magnetic order is present. Harmonic content therefore helps distinguish structures.

A density-wave claim links four levels. Real-space charge or spin modulation defines the operator channel; diffraction satellites identify $\mathbf Q$ and its harmonics; folding mixes $\mathbf k$ and $\mathbf k+\mathbf Q$ and can open a gap $2\lvert\Phi_{\mathbf Q}\rvert$; fluctuations split into amplitude and phase coordinates, with pinning or commensurability giving the phase mode a finite restoring scale.

Commensurate and Incommensurate Order

The wave is commensurate when an integer $m$ exists such that

$$m\mathbf Q=\mathbf G$$

for a reciprocal-lattice vector $\mathbf G$. Then the phase can be locked by an invariant such as

$$F_{\mathrm{lock}} = -g_m \left( \rho_{\mathbf Q}^{\,m} +\rho_{\mathbf Q}^{*m} \right) = -2g_m\lvert\rho_{\mathbf Q}\rvert^m\cos(m\phi).$$

The continuous sliding phase is reduced to $m$ translation-related choices. An incommensurate wave has no finite $m$ and, in an ideal clean continuum description, admits an approximate sliding phase. Real disorder, boundaries, lattice discreteness, and long-range forces pin it. Temperature, pressure, or doping can produce lock-in transitions and discommensurations rather than a smooth change of period.

Nesting

Two Fermi-surface patches are nested by $\mathbf Q$ when translating one by $\mathbf Q$ makes it approximately parallel and coincident with another over an extended region. This enhances the phase space for low-energy particle–hole excitations.

For Bloch bands $n,m$, a bare susceptibility has the form

$$\begin{aligned} \chi_0(\mathbf q,\omega) ={}& \frac{1}{N} \sum_{\mathbf k,n,m} \frac{ f(\epsilon_{n\mathbf k}) - f(\epsilon_{m,\mathbf k+\mathbf q}) }{ \omega+i0^+ +\epsilon_{n\mathbf k} -\epsilon_{m,\mathbf k+\mathbf q} } \\ &\times \left| \mathcal M_{nm}(\mathbf k,\mathbf q) \right|^2 . \end{aligned}$$

The matrix element $\mathcal M_{nm}$ contains orbital and probe information. The low-frequency imaginary part measures near-Fermi-surface joint phase space; the static real part includes virtual transitions over a broader energy range. A geometric nesting picture can therefore disagree with the actual charge, spin, orbital, or phonon-weighted susceptibility.

Enhancement is not an instability

In a selected particle–hole channel $\alpha$, an approximate instability occurs when the largest eigenvalue of the interaction-dressed kernel reaches unity:

$$\lambda_{\max} \left[ \Gamma_{\alpha}(\mathbf q)\, \chi_0(\mathbf q,0) \right] = 1.$$

The vertex $\Gamma_\alpha$ may contain screened Coulomb, exchange, electron–phonon, orbital, and local-field effects. The matrix character matters in a multiorbital crystal. A peak in $\chi_0$ is only one factor; the interaction can favor another wavevector or channel.

For a phonon-coupled CDW, the same statement appears as a softening of a dressed phonon propagator,

$$D^{-1}(\mathbf q,\omega) = D_0^{-1}(\mathbf q,\omega) -\Pi(\mathbf q,\omega).$$

A Kohn anomaly is a nonanalyticity or pronounced softening produced by electronic response. A phonon that reaches zero frequency signals an instability of the assumed high-symmetry structure within the approximation. A broad anomaly that remains finite is evidence for coupling, not proof of static CDW order.

Why nesting is often overclaimed

Real material surfaces are rarely perfectly nested. Curvature, warping, finite lifetime, temperature, spin–orbit coupling, and orbital mismatch cut off the enhancement. In layered dichalcogenides, wavevector-dependent electron–phonon matrix elements and lattice dynamics can be as important as Fermi-surface geometry. In cuprates, charge order emerges from a correlated normal state with pseudogaps and strong momentum-dependent spectral weight, so a bare-band nesting construction is especially incomplete.

A trustworthy mechanism claim should predict the ordering wavevector, transition scale, soft mode or susceptibility evolution, orbital form factor, and response to controlled tuning. Retrofitting one observed $\mathbf Q$ to two visually parallel contours is not enough.

Peierls Instability

The Peierls problem is the controlled ideal in which nesting is strongest. Consider a one-dimensional metal with Fermi points $\pm k_{\mathrm F}$. The static susceptibility has a logarithmic singularity near $2k_{\mathrm F}$:

$$\chi_0(2k_{\mathrm F}+q,0) \sim N(0) \ln \left( \frac{\Lambda}{\lvert v_{\mathrm F}q\rvert} \right).$$

A lattice displacement at $Q=2k_{\mathrm F}$ creates a periodic electronic potential $\Delta_Q$. In the reduced zone, states near $k$ and $k+Q$ mix:

$$\mathcal H_k = \begin{pmatrix} \xi_k & \Delta_Q\\ \Delta_Q^{*} & \xi_{k+Q} \end{pmatrix}.$$

The reconstructed eigenvalues are

$$E_{\pm}(k) = \frac{\xi_k+\xi_{k+Q}}{2} \pm \sqrt{ \left( \frac{\xi_k-\xi_{k+Q}}{2} \right)^2 + \lvert\Delta_Q\rvert^2 }.$$

For perfect nesting, $\xi_{k+Q}=-\xi_k$ and a full gap $2\lvert\Delta_Q\rvert$ opens at the Fermi points. The electronic energy gain contains a logarithm that can overcome the elastic cost for arbitrarily weak coupling in the ideal zero-temperature mean-field limit.

Three qualifications are essential:

1. a strictly one-dimensional fluctuating system is not identical to its static mean-field solution;
2. finite-temperature order in real quasi-one-dimensional compounds relies on interchain coupling, lattice discreteness, long-range interactions, or three-dimensional phonons;
3. imperfect nesting cuts off the logarithm and restores a finite threshold.

The Peierls instability is also different from the Peierls Phase Preview, which concerns the electromagnetic gauge phase attached to a hopping amplitude.

Fermi-Surface Reconstruction

The same $2\times2$ matrix describes the simplest CDW reconstruction in any dimension, with $\Delta_Q$ replaced by a charge-order matrix element $\Phi_{\mathbf Q}(\mathbf k)$. Translation breaking folds the original Brillouin zone, hybridizes states at $\mathbf k$ and $\mathbf k+\mathbf Q$, and opens avoided crossings where their energies are close.

Imperfect nesting usually leaves pockets or arcs of reconstructed quasiparticle contours rather than a full gap. Transport coefficients can change sign, quantum-oscillation frequencies can appear, and optical spectral weight can move from a Drude component into finite-frequency interband transitions. None of these signatures is unique to a density wave; the ordering peak and symmetry information must close the case.

For an SDW, the reconstruction matrix also acts in spin space. A collinear state can mix opposite or equal spin labels depending on the quantization convention, while a spiral mixes spin and momentum simultaneously. Spin–orbit coupling can pin the polarization and gap otherwise soft spin rotations. Itinerant Magnetism develops that matrix susceptibility and magnetic continuum in detail.

Mechanisms Beyond One Cartoon

Several mechanisms can produce the same broken translation symmetry.

Route	Driving physics	Diagnostic emphasis
Peierls or electron–phonon CDW	electronic susceptibility plus a soft lattice mode	phonon dispersion, isotope or pressure response, momentum-dependent coupling
excitonic or electronic CDW	particle–hole attraction or Coulomb-driven coherence	orbital character, collective spectrum, lattice response as secondary or cooperative
local-interaction charge order	intersite repulsion, valence constraints, commensurability	real-space charge disproportionation, strong harmonics, insulating gaps
itinerant SDW	exchange-enhanced finite-$\mathbf Q$ spin susceptibility	magnetic diffraction, spin polarization, reconstructed bands
strong-coupling stripes	kinetic energy competing with local correlations and antiferromagnetism	linked spin and charge wavevectors, domain structure, non-rigid spectral weight

The categories can cooperate. A primarily electronic instability distorts the lattice, and a soft phonon modifies the electronic interaction. “Electronic” and “lattice” are not mutually exclusive once the ordered state is self-consistent.

Collective Modes

Write a single complex density-wave field as

$$\Psi(\mathbf r,t) = \left[ \Psi_0+h(\mathbf r,t) \right] e^{i[\phi_0+\varphi(\mathbf r,t)]}.$$

The radial fluctuation $h$ is an amplitude mode or amplitudon. The angular fluctuation $\varphi$ is a phase mode or phason. In an ideal incommensurate neutral theory, a uniform phase shift translates the wave and costs no energy, so the phason is gapless. Commensurability and disorder add a restoring force and pin it.

Charge order couples the phason to electric fields, normal carriers, and long-range Coulomb forces. Screening, dimensionality, and pinning determine whether the mode remains low, becomes plasmon-like, or appears as a broad relaxational response. Above the transition, a soft phonon, central peak, or overdamped order-parameter fluctuation may replace a sharp collective pole.

An SDW has additional orientation modes. If spin rotation is broken continuously, transverse spin waves appear; spin–orbit anisotropy gaps or splits them. Longitudinal amplitude fluctuations usually overlap a particle–hole continuum and can be strongly damped. A sharp Raman line or pump–probe oscillation should therefore be assigned only after symmetry, polarization, temperature, and continuum coupling are checked.

Sliding transport

An incommensurate CDW can depin above a threshold electric field and contribute nonlinear collective current. Narrow-band noise, mode locking under combined dc and ac drive, and hysteretic threshold behavior are classic evidence. This is not ordinary zero-resistance superconductivity: pinning, dissipation through normal carriers, contacts, and phase slips remain central.

Competition and Coexistence with Superconductivity

Let $\rho$ denote a density-wave amplitude and $\Delta$ a uniform superconducting order parameter. The lowest local coupling consistent with both symmetries is

$$\begin{aligned} F ={}& a_{\rho}\lvert\rho\rvert^2 + \frac{b_{\rho}}{2}\lvert\rho\rvert^4 + a_{\Delta}\lvert\Delta\rvert^2 + \frac{b_{\Delta}}{2}\lvert\Delta\rvert^4 \\ &+ \gamma \lvert\rho\rvert^2 \lvert\Delta\rvert^2 . \end{aligned}$$

A positive $\gamma$ expresses local competition; a negative $\gamma$ favors mutual enhancement. With $b_{\rho},b_{\Delta}>0$, a locally stable interior coexistence solution requires

$$b_{\rho}b_{\Delta}-\gamma^2>0$$

and the resulting squared amplitudes must both be positive. This determinant condition is stronger than boundedness when $\gamma$ is large and positive: because the physical variables $\lvert\rho\rvert^2$ and $\lvert\Delta\rvert^2$ are nonnegative, a large positive cross-coupling can leave the free energy bounded while favoring an axis minimum with only one order. Thus competition does not imply mutual exclusion. Spatial inhomogeneity, disorder, gradients, multiple wavevectors, and magnetic field can produce coexistence in different regions even when the local coupling is repulsive.

Microscopically, the two orders often compete for the same low-energy states. A density-wave gap reduces the density of states available for pairing; superconductivity can in turn weaken the diffraction peak below $T_c$. Conversely, a soft density-wave mode can enhance an effective pairing interaction, and partial reconstruction can leave pockets that still pair.

In underdoped YBa$_2$Cu$_3$O$_{6+x}$, x-ray measurements show charge order that is weakened below $T_c$ and strengthened when magnetic field suppresses superconductivity. This is direct evidence for competition in that material and regime, not a universal sign of $\gamma$ for every density wave. Stripe-ordered La-based cuprates demonstrate that spin order, charge order, lattice symmetry, and superconducting coherence can be intertwined in still more structured ways.

A pair-density wave is a spatial modulation of the pair field, not merely a CDW coexisting with uniform superconductivity. Couplings can induce secondary charge harmonics, but the pair order requires a pair-sensitive diagnostic.

Experimental Evidence

Probe	Primary sensitivity	Strong evidence	Main caveat
nonresonant X-ray diffraction	atomic displacements and total charge	satellites at $\mathbf G\pm\mathbf Q$, correlation length, harmonics	intensity may be dominated by the lattice distortion
resonant X-ray scattering	element- and orbital-selective electronic response	energy and polarization dependence tied to a form factor	absorption, self-absorption, and tensor modeling
neutron diffraction	magnetic moment and nuclear structure	magnetic satellites, polarization, spin orientation	small moments and finite time window
ARPES	occupied single-particle spectrum	folded bands, avoided crossings, coherence factors	matrix elements, surfaces, domains, pseudogaps
STM and spectroscopy	surface local density and tunneling conductance	real-space phase, defects, form-factor-resolved Fourier peaks	surface selection and setpoint effects
optical and transport probes	current response and reconstructed carriers	spectral-weight transfer, collective pinning, nonlinear sliding	gaps and anomalies are not symmetry-specific
NMR, NQR, and muon probes	local charge or magnetic environment	line splitting, internal fields, fluctuation timescales	local disorder can mimic broad distributions

The minimum static claim should report $\mathbf Q$, peak width, temperature dependence, background subtraction, instrumental resolution, and whether the signal scales as a bulk correlation volume. Mechanism claims require more: the soft mode or susceptibility that leads to the transition, orbital and phonon content, and response to at least one controlled tuning variable.

Evidence ladder

1. Tendency: a susceptibility peak, softened phonon, or enhanced finite-$\mathbf Q$ correlations.
2. Quasi-static correlations: a finite-width peak within the probe’s time window.
3. Long-range order: resolution-limited or scaling-supported correlations with a symmetry-consistent order parameter.
4. Reconstruction: folded bands or transport changes quantitatively tied to the same $\mathbf Q$.
5. Mechanism: one microscopic model explains the transition, collective modes, form factors, and tuning response.

Skipping from level 1 to level 5 is the most common overclaim.

Material Examples

Quasi-one-dimensional conductors

Compounds such as blue bronze and NbSe$_3$ display multiple classic CDW signatures: superlattice peaks, partial electronic gaps, pinned collective response, threshold conduction, and narrow-band noise. Their anisotropy makes the Peierls language useful, while interchain coupling and disorder explain why the real transition is not a strictly one-dimensional mean-field theorem.

Layered transition-metal dichalcogenides

Materials such as 2H-NbSe$_2$ combine charge order, soft phonons, multiband electronic structure, and superconductivity. The ordering wavevector is not explained reliably by Fermi-surface geometry alone; momentum-dependent electron–phonon coupling and anharmonic lattice dynamics are part of the mechanism.

Chromium

Chromium is the canonical itinerant SDW metal. Incommensurate magnetic order, a concomitant charge and strain modulation, reconstructed electronic structure, and pressure or alloy dependence can be related to electron and hole Fermi-surface sheets. Even here, quantitative theory requires more than a nesting cartoon.

Cuprates

Cuprate charge order and spin stripes live in a strongly correlated, often pseudogapped background. Their wavevectors, orbital form factors, short correlation lengths, field response, and relation to superconductivity vary across families. The t–J Model supplies one projected strong-coupling language, but the observed density wave is not thereby reduced to weak-coupling nesting or to one universal stripe mechanism.

Common Mistakes

Calling any Fourier peak an order parameter

Bragg peaks from the original lattice, quasiparticle-interference peaks, impurity Friedel oscillations, and finite-range correlations can all appear in Fourier space. Temperature, field, energy, phase coherence, and size scaling distinguish them.

Equating a partial gap with a CDW

Superconductivity, hybridization, magnetism, Mott physics, disorder, and matrix-element suppression can all reduce spectral weight. A density-wave gap must be tied to translation breaking and the same $\mathbf Q$ that folds the states.

Treating nesting as sufficient

The relevant susceptibility includes orbital matrix elements and a channel-dependent vertex. The strongest geometric nesting vector can differ from the observed order, and many observed CDWs are cooperative electronic–lattice instabilities.

Calling every phonon anomaly Peierls

A Kohn anomaly records electronic screening of a phonon. The Peierls limit additionally requires a one-dimensional $2k_{\mathrm F}$ singularity and a self-consistent symmetry-breaking distortion.

Confusing the Peierls instability and Peierls phase

The instability creates a density wave through electron–lattice coupling. The Peierls phase is a gauge-covariant phase factor on hopping in an electromagnetic vector potential.

Treating finite probe time as truly static

A peak can look elastic when fluctuations are slower than instrumental resolution. Comparing neutron, x-ray, NMR, muon, and pump–probe windows can reveal glassy or slowly fluctuating order.

Assuming competition forbids coexistence

A positive local coupling can still permit coexistence when both bare tendencies are strong and the quartic form is stable. Domains, vortices, and disorder create further spatial coexistence.

Exercises

1. Translation and time-reversal audit

For a scalar CDW amplitude $\rho_{\mathbf Q}$ and a collinear SDW amplitude $\mathbf M_{\mathbf Q}$, determine their transformation under translation by $\mathbf a$ and time reversal. Which symmetries are necessarily broken by a generic incommensurate state?

Solution

Translation gives the phase factors

$$\rho_{\mathbf Q} \mapsto e^{i\mathbf Q\cdot\mathbf a}\rho_{\mathbf Q}, \qquad \mathbf M_{\mathbf Q} \mapsto e^{i\mathbf Q\cdot\mathbf a}\mathbf M_{\mathbf Q}.$$

Time reversal acts on the physical profiles without reflecting space:

$$\rho_{\mathbf Q}\mapsto\rho_{\mathbf Q}, \qquad \mathbf M_{\mathbf Q}\mapsto-\mathbf M_{\mathbf Q}.$$

The operator identities $\mathcal T\widehat\rho_{\mathbf Q}\mathcal T^{-1}=\widehat\rho_{-\mathbf Q}$ and $\mathcal T\widehat{\mathbf S}_{\mathbf Q}\mathcal T^{-1}=-\widehat{\mathbf S}_{-\mathbf Q}$ contain a conjugated wavevector because $\mathcal T$ is antiunitary. The transformed expectation value is conjugated once more. The reality relation $\rho_{-\mathbf Q}=\rho_{\mathbf Q}^{*}$ should therefore not be mistaken for a time-reversal shift of the physical charge pattern.

A generic incommensurate CDW breaks lattice translation but need not break time reversal. A collinear SDW breaks translation and time reversal; without spin–orbit pinning it also breaks continuous spin rotation down to rotations about the ordered axis. Point-group symmetry may also break if one member of a symmetry-related star of wavevectors is selected.

2. Density-wave reconstruction

Diagonalize

$$\mathcal H_k = \begin{pmatrix} \xi_k & \Delta\\ \Delta^{*} & \xi_{k+Q} \end{pmatrix}$$

and show that perfect nesting produces a gap $2\lvert\Delta\rvert$.

Solution

The characteristic equation gives

$$E_{\pm} = \frac{\xi_k+\xi_{k+Q}}{2} \pm \sqrt{ \left( \frac{\xi_k-\xi_{k+Q}}{2} \right)^2 + \lvert\Delta\rvert^2 }.$$

Under perfect nesting, $\xi_{k+Q}=-\xi_k$, so

$$E_{\pm} = \pm\sqrt{\xi_k^2+\lvert\Delta\rvert^2}.$$

At the original Fermi crossing $\xi_k=0$, the upper and lower branches lie at $\pm\lvert\Delta\rvert$. Their separation is therefore $2\lvert\Delta\rvert$.

3. Cutoff of the Peierls singularity

The ideal one-dimensional susceptibility contains

$$\chi_0(2k_{\mathrm F}+q,0) \sim N(0)\ln \left( \frac{\Lambda}{\lvert v_{\mathrm F}q\rvert} \right).$$

Give three physical effects that replace the vanishing denominator by a finite scale, and state their consequence for the instability.

Solution

Finite temperature smears the Fermi edge and supplies a scale of order $k_{\mathrm B}T$. A finite quasiparticle scattering rate supplies $\hbar/\tau$. Transverse hopping or Fermi-surface curvature produces an imperfect-nesting mismatch $\delta_{\mathrm{nest}}$. Schematically,

$$\lvert v_{\mathrm F}q\rvert \longrightarrow \max \left( \lvert v_{\mathrm F}q\rvert, k_{\mathrm B}T, \frac{\hbar}{\tau}, \delta_{\mathrm{nest}} \right).$$

The logarithm then remains finite. An arbitrarily weak coupling no longer guarantees order; the interaction and lattice gain must exceed a finite threshold set by the cutoff and elastic cost.

4. Charge harmonic induced by a collinear SDW

Let $\mathbf S(\mathbf r)=\mathbf M_0\cos(\mathbf Q\cdot\mathbf r+\phi)$. Show that the scalar $\mathbf S^2$ contains a component at $2\mathbf Q$, and explain why the same conclusion can fail for a circular spiral.

Solution

Using $\cos^2\theta=(1+\cos2\theta)/2$,

$$\mathbf S^2(\mathbf r) = \frac{\lvert\mathbf M_0\rvert^2}{2} \left[ 1+ \cos(2\mathbf Q\cdot\mathbf r+2\phi) \right].$$

A scalar charge or lattice field can therefore couple linearly to the $2\mathbf Q$ harmonic.

For a circular spiral, write

$$\mathbf M_{\mathbf Q} = \frac{M}{2} \left( \widehat{\mathbf x} -i\widehat{\mathbf y} \right).$$

Then $\mathbf M_{\mathbf Q}\cdot\mathbf M_{\mathbf Q}=0$. The spin magnitude is spatially constant, so the simplest quadratic mechanism does not induce a $2\mathbf Q$ charge modulation.

5. Homogeneous coexistence with superconductivity

For the Landau free energy above, set $x=\lvert\rho\rvert^2$ and $y=\lvert\Delta\rvert^2$. Solve for a stationary coexistence state and state the stability and positivity conditions.

Solution

The stationarity equations are

$$\begin{pmatrix} b_{\rho} & \gamma\\ \gamma & b_{\Delta} \end{pmatrix} \begin{pmatrix} x\\y \end{pmatrix} = - \begin{pmatrix} a_{\rho}\\a_{\Delta} \end{pmatrix}.$$

Inverting gives

$$\begin{aligned} x &= \frac{ -a_{\rho}b_{\Delta} +\gamma a_{\Delta} }{ b_{\rho}b_{\Delta}-\gamma^2 }, \\ y &= \frac{ -a_{\Delta}b_{\rho} +\gamma a_{\rho} }{ b_{\rho}b_{\Delta}-\gamma^2 }. \end{aligned}$$

The Hessian in the interior variables $(x,y)$ is positive definite when $b_{\rho}>0$, $b_{\Delta}>0$, and $b_{\rho}b_{\Delta}>\gamma^2$. Homogeneous coexistence additionally requires $x>0$ and $y>0$. A positive $\gamma$ suppresses both amplitudes relative to uncoupled values but does not automatically make either vanish. If $\gamma>\sqrt{b_{\rho}b_{\Delta}}$, the quartic energy is still bounded on the physical quadrant $x,y\geq0$, but the interior stationary point is not a coexistence minimum.

6. Peak growth versus long-range order

For $N=L^d$ sites, define

$$S(\mathbf Q) = \frac{1}{N} \sum_{i,j} e^{i\mathbf Q\cdot(\mathbf r_i-\mathbf r_j)} \langle O_i O_j\rangle.$$

How does $S(\mathbf Q)$ scale for true long-range order, finite-correlation-length order, and algebraic correlations $\langle O(\mathbf r)O(0)\rangle\sim r^{-\eta}$?

Solution

If the modulated correlation approaches $m^2$ at long distance, the double sum contains $N^2m^2$ coherent terms and the prefactor leaves

$$S(\mathbf Q)\sim Nm^2.$$

Thus $S(\mathbf Q)/N\to m^2$.

For finite correlation length $\xi$, each site correlates coherently with only a volume of order $\xi^d$, so $S(\mathbf Q)$ approaches a size-independent value proportional to $\xi^d$ once $L\gg\xi$.

For algebraic correlations, integrate $r^{-\eta}$ over a volume of radius $L$:

$$S(\mathbf Q) \sim L^{d-\eta}$$

for $0<\eta<d$. The peak diverges but remains subextensive, so $S(\mathbf Q)/N\sim L^{-\eta}\to0$. A growing raw peak is therefore not sufficient evidence for long-range order.

Research Status

The symmetry classification, Peierls limit, response-function framework, band folding, commensurability, and amplitude/phase collective coordinates are standard. CDWs in quasi-one-dimensional conductors and SDW order in chromium are mature benchmark subjects.

Material mechanisms remain system-dependent. In many layered compounds, nesting, electron–phonon coupling, excitonic effects, anharmonicity, orbital texture, and correlations cooperate. In cuprates and other strongly correlated systems, the origin, dimensionality, fluctuation spectrum, and relationship of charge order to pseudogap and superconductivity remain active. Claims should distinguish an observed order parameter from a proposed microscopic driver.

Connections

• Order Parameters — symmetry transformation, source fields, and density-wave operator construction.
• Long-Range Order — correlation limits and finite-size scaling.
• Structure Factors — elastic and dynamical momentum-resolved response.
• Particle–Hole Excitations — continua and low-energy phase space.
• Fermi Surface — nesting geometry and reconstruction.
• Random Phase Approximation — matrix susceptibility and pole criteria.
• Itinerant Magnetism — finite-$\mathbf Q$ magnetic instability and chromium.
• Antiferromagnetism — local and itinerant finite-wavevector magnetic order.
• Phonons — normal modes, softening, linewidths, and lattice dynamics.
• BCS Theory — the uniform weak-coupling superconducting baseline.
• Competing Orders — coupled-order phase topology and evidence standards for density waves near superconducting, magnetic, or nematic phases.
• Pair-Density Waves and Exotic Orders — the distinct finite-momentum pair operator, induced charge harmonics, phase defects, and pair-sensitive diagnostics.
• t–J Model — projected strong-coupling routes to stripes and competing correlations.
• Transition-Metal Dichalcogenides — the semiconducting monolayer and moiré branch of the TMD family, distinct from the metallic-TMD density waves treated here.

References

• W. Kohn, “Image of the Fermi surface in the vibration spectrum of a metal,” Physical Review Letters 2, 393–394 (1959).
• A. W. Overhauser, “Spin density waves in an electron gas,” Physical Review 128, 1437–1452 (1962).
• G. Grüner, “The dynamics of charge-density waves,” Reviews of Modern Physics 60, 1129–1181 (1988).
• E. Fawcett, “Spin-density-wave antiferromagnetism in chromium,” Reviews of Modern Physics 60, 209–283 (1988).
• P. A. Lee, T. M. Rice, and P. W. Anderson, “Fluctuation effects at a Peierls transition,” Solid State Communications 14, 703–709 (1974).
• M. D. Johannes and I. I. Mazin, “Fermi surface nesting and the origin of charge density waves in metals,” Physical Review B 77, 165135 (2008).
• F. Weber, S. Rosenkranz, J.-P. Castellan, et al., “Extended phonon collapse and the origin of the charge-density wave in 2H-NbSe$_2$,” Physical Review Letters 107, 107403 (2011).
• P. B. Littlewood and C. M. Varma, “Amplitude collective modes in superconductors and their coupling to charge-density waves,” Physical Review B 26, 4883–4893 (1982).
• J. A. Wilson, F. J. Di Salvo, and S. Mahajan, “Charge-density waves and superlattices in the metallic layered transition metal dichalcogenides,” Advances in Physics 24, 117–201 (1975).
• P. Monceau, “Electronic crystals: An experimental overview,” Advances in Physics 61, 325–581 (2012).
• J. M. Tranquada, B. J. Sternlieb, J. D. Axe, Y. Nakamura, and S. Uchida, “Evidence for stripe correlations of spins and holes in copper oxide superconductors,” Nature 375, 561–563 (1995).
• J. Chang, E. Blackburn, A. T. Holmes, et al., “Direct observation of competition between superconductivity and charge density wave order in YBa$_2$Cu$_3$O$_{6.67}$,” Nature Physics 8, 871–876 (2012).
• E. Fradkin, S. A. Kivelson, and J. M. Tranquada, “Colloquium: Theory of intertwined orders in high temperature superconductors,” Reviews of Modern Physics 87, 457–482 (2015).
• R. Comin and A. Damascelli, “Resonant x-ray scattering studies of charge order in cuprates,” Annual Review of Condensed Matter Physics 7, 369–405 (2016).
• M. D. Johannes, I. I. Mazin, and C. A. Howells, “Fermi-surface nesting and the origin of the charge-density wave in NbSe$_2$,” Physical Review B 73, 205102 (2006).

Summary

Charge and spin density waves are finite-wavevector ordered states defined by an operator, symmetry, wavevector, and thermodynamic correlation criterion. Nesting enhances particle–hole phase space but becomes an instability only after channel-dependent interactions and matrix elements are included. The Peierls model supplies an ideal one-dimensional limit; real materials add imperfect nesting, phonons, orbitals, disorder, dimensional crossover, or strong coupling. Ordered waves reconstruct bands and support amplitude, phase, and spin-orientation modes. Their relationship with superconductivity can be competitive, cooperative, homogeneous, or spatially intertwined, and must be established with cross-probe evidence rather than inferred from one gap or one Fourier peak.