---
title: Quasi-1D Wannier Excitons
url: https://www.emergentmind.com/topics/quasi-one-dimensional-wannier-excitons
type: topic
---

# Quasi-1D Wannier Excitons

Searching arXiv for recent and foundational papers on quasi-one-dimensional Wannier excitons.
Quasi-one-dimensional Wannier excitons are bound electron–hole states whose relative motion is effectively restricted to a single spatial direction because the transverse motion is either frozen out by strong confinement or rendered subdominant by strong in-plane anisotropy. In this regime, the Coulomb interaction is commonly represented by regularized one-dimensional kernels, the screening is weak, and the resulting binding can range from weakly bound holon–doublon excitons in a one-dimensional Mott insulator to strongly bound yet still Wannier-Mott–like excitons in atomic chains, nanotubes, and anisotropic two-dimensional crystals. A defining point emerging across recent work is that strong binding in reduced dimensionality does not by itself imply Frenkel localization: in several quasi-1D systems the average electron–hole separation remains larger than the lattice constant, preserving Wannier character [2006.09004; 1405.0777; 1411.6124; 2510.09194].

## 1. Dimensional reduction and effective Hamiltonians

In quasi-1D semiconductors such as nanowires and carbon nanotubes, the standard starting point is a geometry so strongly confined in the two transverse directions that only the lowest subband remains. The fast transverse motion is then “frozen out,” and the electron–hole coordinate along the axial direction is treated quantum-mechanically. A common effective attraction is the one-dimensional “cusp” potential
\[
V(z)=-\frac{1}{|z|+z_0},
\]
where \(z_0>0\) is a short-range cutoff whose order of magnitude is set by the transverse confinement radius \(R\) [1405.0777]. In carbon nanotubes, the same effective-mass picture leads to
\[
\hat H_X=-\frac{\hbar^2}{2\mu}\frac{d^2}{dz^2}-\frac{e^2}{4\pi\epsilon}\frac{1}{|z|+z_0},
\]
with \(z=z_e-z_h\), \(\mu=m/2\) for equal electron and hole masses, and \(z_0\sim r\) regularizing the one-dimensional divergence at \(z=0\) [1010.6035].

Phosphorene realizes a different route to quasi-one-dimensionality. Because it is atomically thin but strongly anisotropic, with much lighter masses along the armchair or \(x\)-direction than along the zigzag or \(y\)-direction, its exciton can be described as a two-dimensional carrier pair that becomes bound in an effectively one-dimensional relative coordinate. In the effective-mass approximation, the relative motion along the strongly confined direction satisfies
\[
\left[-\frac{\hbar^2}{2\mu_x}\frac{d^2}{dx^2}+V_{1D}(x)\right]\psi(x)=E_b\psi(x),
\]
with \(\mu_x^{-1}=m_{e,x}^{-1}+m_{h,x}^{-1}\) and a regularized interaction that may be approximated by
\[
V_{1D}(x)\approx -\,\frac{e^2}{4\pi\varepsilon_0\kappa}\,\frac{1}{\sqrt{x^2+r_0^2}},
\]
where \(r_0\) is a screening length on the order of a few nanometers [1411.6124].

A more microscopic formulation for truly one-dimensional atomic chains is the Bethe–Salpeter equation,
\[
\sum_{v'c'k'}\Bigl[(E^{\rm GW}_{ck}-E^{\rm GW}_{vk})\delta_{vv'}\delta_{cc'}\delta_{kk'}+K^{\rm eh}_{vck,v'c'k'}\Bigr]A^S_{v'c'k'}=E^S A^S_{vck},
\]
with the exciton binding energy defined by
\[
E_b=E_g^{\rm GW}-E_{\rm opt}^{\rm BSE}.
\]
This form is used for freestanding exfoliable one-dimensional chains of S\(_3\), Te\(_3\), As\(_2\)S\(_3\), and Bi\(_2\)Te\(_3\) [2510.09194].

Quasi-1D Wannier excitons also arise in correlated-electron models. For a one-dimensional extended Hubbard model at half-filling with a periodic modulated field, a spin-charge-separation treatment projects onto sectors with \(M\) holon–doublon pairs,
\[
H^{(\mathrm C)}=\sum_{M,M'}P_M\,(H_0+H_V+H_\phi)\,P_{M'},
\]
yielding a charge model whose basis states keep the remaining \(N-2M\) sites in a Heisenberg ground state. In that setting, the exciton is a many-body holon–doublon bound state rather than a simple band exciton [2006.09004].

## 2. Binding energies, spatial extent, and the Wannier–Mott criterion

The quasi-1D regime supports a broad range of exciton binding energies, from weakly bound states in correlated Mott systems to multi-electron-volt excitons in isolated atomic chains. The central physical distinction is not only the magnitude of \(E_b\), but whether the real-space extent exceeds the microscopic lattice scale.

| Platform | Reported binding/energy scales | Spatial character |
|---|---:|---|
| ET-F\(_2\)TCNQ charge model | \(E_b^-\approx0.16T\approx0.017\) eV for \(V_1=2.8T\); \(E_b^-\approx0.04\) eV and \(E_b^+\approx0.02\) eV for the long-range fit | \(\psi_1(r)\) decays over \(r\sim5\) lattice sites |
| Semiconducting carbon nanotubes | \(E_{X*}\approx150\)–200 meV and \(E_{XX}\approx100\)–130 meV at \(d\sim0.7\)–1.1 nm | trion-to-biexciton ratio \(R\approx1.3\)–1.5 for small diameter |
| Few-layer phosphorene | \(E_b^{\rm ex}(N)=A/(N^\alpha+B)\), with \(A=0.83\) eV, \(\alpha=0.53\), \(B=0.03\) eV; monolayer \(E_b^{\rm ex}\approx0.86\) eV; trion \(\Delta E\approx180\)–190 meV in 2L–3L | completely polarized armchair emission demonstrates quasi-1D confinement |
| S\(_3\), Te\(_3\), As\(_2\)S\(_3\), Bi\(_2\)Te\(_3\) chains | \(E_b=2.27,\,2.14,\,0.33,\,0.85\) eV | \(r_{\rm eh}=12.4,\,14.7,\,14.4,\,13.2\) Å, all larger than \(a\) |

In ET-F\(_2\)TCNQ, the Mott continuum threshold \(E_{\rm gap}\equiv\omega_{\rm cont}\) and exciton energy \(E_{\rm exc}\equiv\omega_1\) define
\[
E_b=E_{\rm gap}-E_{\rm exc}.
\]
For the fit \(V_1=2.8T\), \(V_{\alpha\ge2}=0\), \(T=0.107\) eV, the odd-parity exciton lies at \(\widetilde E^-_{\rm exc}=6.426T\approx0.687\) eV, the continuum onset is \(\approx6.59T\approx0.706\) eV, and the binding is \(E_b^-\approx0.16T\approx0.017\) eV, while the lowest even-parity level is unbound. For the longer-range interaction \(V_1=2.4T\), \(V_2=V/2\), \(V_3=V/3\), \(T=0.111\) eV, both odd- and even-parity excitons are weakly bound, but \(E_b^+<E_b^-\) [2006.09004].

The same dimensional reduction can, however, produce much stronger binding without destroying Wannier character. In the exfoliable one-dimensional chains, the average electron–hole separations exceed the lattice constants in every case, which confirms a Wannier-Mott–like character even though the binding energies reach \(2.27\) eV in S\(_3\) and \(2.14\) eV in Te\(_3\) [2510.09194]. This directly counters the common assumption that an exciton with an electron-volt-scale binding energy in 1D must be Frenkel-like.

Phosphorene occupies an intermediate position. Its quasi-1D excitonic behavior arises in a higher-than-one-dimensional material, with monolayer \(E_b^{\rm ex}\approx0.86\) eV and room-temperature trion binding energies near \(190\) meV in few-layer samples [1411.6124]. By contrast, the Mott-insulator study finds that many-body corrections \(\Delta\widetilde E^\pm\) renormalize the effective attraction \(V\to\widetilde V^\pm\), reducing binding by \(\sim30\%\), so that \(E_b\ll T\), typically \(10\)–\(50\) meV [2006.09004]. A plausible implication is that “quasi-one-dimensional Wannier exciton” denotes a common kinematic regime, not a universal binding scale.

## 3. Optical spectra, parity, and momentum-space signatures

The most direct spectroscopic signatures of quasi-1D Wannier excitons depend strongly on symmetry. In the one-dimensional Mott-insulator charge model, once the exciton eigenpairs \(\{\omega_\lambda,|\psi_\lambda\rangle\}\) and transition vector \(|J_{\rm ex}\rangle\) are known, the optical conductivity at \(T=0\) is
\[
\sigma(\omega)=\frac{\gamma}{\omega}\sum_\lambda
\frac{|\langle\psi_\lambda|J_{\rm ex}\rangle|^2}{(\omega-\omega_\lambda)^2+\gamma^2}.
\]
Only odd-parity excitons carry nonzero \(\langle\psi_\lambda|J_{\rm ex}\rangle\), so the odd state is dipole-allowed in linear absorption, whereas the even-parity exciton is dark in \(\sigma(\omega)\) but can be mixed in by the modulated field \(H_\phi\). The plus–minus–plus feature in \(\Delta\sigma(\omega)\equiv \sigma|_{\varepsilon\neq0}-\sigma|_{\varepsilon=0}\) arises from field-mixing of even and odd states via \(H_\phi\) [2006.09004]. A recurring misconception is that a dark exciton is absent from the spectrum; in this case it is present in the eigenvalue structure but optically inactive in the unmodulated linear-response channel.

Phosphorene provides an optical signature of quasi-1D confinement that is not based on parity but on in-plane polarization. Gate-dependent photoluminescence shows two Lorentz-shaped peaks corresponding to the neutral exciton \(X\) and the charged trion \(T\), with \(\Delta E\approx E_X-E_T\) interpreted as the trion binding energy. The emission polarization follows \(I(\theta_2)\propto\cos^2(\theta_2-\theta_0)\), with linear dichroism \(LD\approx1.0\) for excitons and \(\approx0.9\) for trions. The fact that the emission is completely polarized along the armchair direction, independent of excitation, demonstrates that the exciton is confined to move and recombine in the \(x\)-direction only, i.e. it is quasi-1D [1411.6124].

In isolated one-dimensional chains, BSE spectra exhibit pronounced bound-exciton peaks well below the GW gaps. The first bright excitons occur at \(4.07\) eV in S\(_3\), \(2.17\) eV in Te\(_3\), \(2.71\), \(2.75\), and \(2.79\) eV in As\(_2\)S\(_3\), and \(0.80\) eV in Bi\(_2\)Te\(_3\), all for light polarized along the chain. Each feature lies well below the corresponding GW gap and carries an oscillator strength typical of Wannier excitons [2510.09194].

A qualitatively different probe is angle-resolved photoemission in quasi-one-dimensional metallic TaSe\(_3\). There, two and under stronger doping three side-valence bands disperse nearly parallel to the main valence band but are shifted to lower excitation energies by momentum-independent offsets. In the pristine sample at \(15\) K, the first side band sits about \(\Delta_1\simeq70\) meV above the main valence band and the second at \(\Delta_2\simeq37\) meV; after moderate K-doping, \(\Delta_1\) increases to \(\simeq180\) meV and \(\Delta_2\) to \(\simeq70\) meV. These thresholds are interpreted as mobile intrachain and interchain excitons, and possibly trions, thereby extending quasi-1D exciton spectroscopy to finite-momentum, itinerant bound states [2009.07157].

## 4. Excitonic complexes and exchange physics

Quasi-1D confinement changes not only the lowest neutral exciton but also the relative stability of excitonic complexes. A configuration-space approach for neutral biexcitons \(XX\) and charged trions \(X^\*\) gives the analytic asymptotic results
\[
E_{XX}= -\frac{1}{9}|E_X|\left(\frac{e}{3}\right)^{2\sqrt{Ry^\*/|E_X|}-1},
\]
and
\[
E_{X^\*}= -\frac{|E_X|}{e\,2^{2\sqrt{Ry^\*/|E_X|}}}.
\]
The ratio
\[
R\equiv \frac{|E_{X^\*}|}{|E_{XX}|}
=3\left(\frac{3}{2e}\right)^{2\sqrt{Ry^\*/|E_X|}}
\]
determines stability: \(R>1\) implies the trion is more stable than the biexciton, whereas \(R<1\) implies the reverse. In terms of \(\nu_0=\sqrt{Ry^\*/|E_X|}\), the critical value is \(\nu_{0,\mathrm{crit}}\simeq0.92\) [1405.0777].

This criterion has a direct carbon-nanotube application. Using the empirical scaling \(|E_X|\approx Ry^\*/r^{0.6}\), with \(r=R/a_B^\*\), one obtains \(\nu_0=r^{0.3}\) and therefore \(R=3(3/2e)^{2r^{0.3}}\). For small-diameter tubes \(r\lesssim0.3\)–\(0.5\), one finds \(R\approx1.3\)–\(1.5\), in good agreement with experiment, while increasing diameter decreases \(R\) and eventually drives the biexciton to greater stability [1405.0777]. A common overgeneralization is that charged complexes are always less stable than neutral ones; quasi-1D systems violate that expectation in the strongly confined, small-\(\mu\) regime.

The same physics can be phrased as asymptotic tunnel exchange. For two quasi-1D excitons in a carbon nanotube separated by \(\Delta Z\), the tunnel integral is
\[
J(\Delta Z)=\frac{2}{3\nu_0^3}\left(\frac{e}{3}\right)^{2\nu_0}\Delta Z\,\exp\!\left[-\frac{2\Delta Z}{\nu_0}\right],
\qquad \Delta Z\gg1,
\]
and the biexciton binding energy is \(E_{XX}(\Delta Z)=-J(\Delta Z)\). The optimum separation \(\Delta Z_0=\nu_0/2\) gives the equilibrium binding
\[
E_{XX}= -\,\frac{1}{9\,\nu_0^2}\left(\frac{e}{3}\right)^{2\nu_0-1}
= -\,\frac{1}{9}|E_X|\left(\frac{e}{3}\right)^{2\sqrt{Ry^\*/|E_X|}-1}.
\]
In the strong-confinement limit, \(|E_{XX}|/|E_X|\to 1/(3e)\approx0.12\) [1010.6035]. This exchange formulation makes explicit that the biexciton is controlled by tunneling between equivalent minima in the two-exciton configuration space.

## 5. Static electric fields, Stark shifts, and exciton dissociation

A notable peculiarity of the one-dimensional problem is that in the absence of static field the electron and hole are always bound, forming an exciton regardless of the Coulomb interaction strength. In a two-band lattice model with one conduction-band orbital and one valence-band orbital per site, the relative-coordinate equation at \(K=0\) is
\[
E\,\phi_\ell=(F\ell-U)\phi_\ell-T(\phi_{\ell+1}+\phi_{\ell-1}),
\qquad U\equiv U_0+2V_d,\quad T\equiv t_e+t_h,
\]
where \(F=e\mathcal E a\) is the field in energy-per-site units. At \(F=0\), the bound-state solution is
\[
\phi_\ell^{(0)}=\sqrt{\tanh\alpha}\,e^{-\alpha|\ell|},
\qquad
e^{-\alpha}=\frac{\sqrt{U^2+4T^2}-U}{2T},
\]
with energy
\[
E_0=-\sqrt{U^2+4T^2},
\]
and binding energy
\[
E_b=\sqrt{U^2+4T^2}-2T.
\]
At low static fields, the main exciton absorption line exhibits a quadratic redshift, \(\Delta E(F)\approx-\kappa_s(U)F^2\); when the field exceeds a critical threshold, the exciton dissociates and the linear optical spectra exhibit the Wannier–Stark ladder with equally spaced peaks [2507.17337].

Above the dissociation scale, the noninteracting solutions are
\[
\phi_\ell(\nu)=J_{\ell-\nu}(2T/F),\qquad E_\nu=\nu F,
\]
so the ladder spacing is \(\Delta E=F=e\mathcal E a\) [2507.17337]. Spectroscopically, the field-dependent optical response evolves from a single bound exciton peak through a broadened, quenched resonance into a fan of Stark lines. This provides a direct experimental fingerprint of exciton ionization in quasi-1D semiconductors.

The continuum Wannier model in a uniform field admits an exact Airy-function treatment. In dimensionless atomic units, the Hamiltonian is
\[
H\phi(x)=\left[-\frac12\frac{d^2}{dx^2}-Z\delta(x)+Fx\right]\phi(x)=E\phi(x),
\]
with \(Z>0\) the \(\delta\)-strength and \(F>0\) the dimensionless field. Away from \(x=0\), the solution is built from Airy functions and the outgoing-wave resonance condition is
\[
Z\,Ai(-\zeta)\,\bigl[Ai(-\zeta)-i\,Bi(-\zeta)\bigr]+p=0,
\]
where \(p=(2F)^{1/3}\) and \(\zeta=-E/p\). A complex resonance \(E(F)=\mathrm{Re}\,E-(i/2)\Gamma\) yields both the Stark shift and the ionization rate. The small-field expansion begins
\[
E(F)= -\frac12 Z^2-\frac{5}{32Z^4}F^2-\frac{55}{8710Z^6}F^4-\cdots,
\]
and the ionization rate has the hydrogen-like tunneling form
\[
\Gamma(F)\propto F^2\exp\!\left[-\frac{4Z^3}{3F}\right]
\]
for \(F\to0\). The same framework gives closed-form expressions for exciton electroabsorption and dynamic polarizability [2601.09272].

## 6. Computational formalisms and localized exciton representations

Several complementary computational frameworks are now used to describe quasi-1D Wannier excitons. In the one-dimensional Mott-insulator problem, the many-body Wannier functions method begins by solving the \(N=18\) charge model in each parity sector at center-of-mass momentum \(K=0\),
\[
H_0^{(\mathrm C)}|\Phi_\mu^p(0)\rangle=E_\mu^p|\Phi_\mu^p(0)\rangle.
\]
Within the one-pair subspace, a unitary Wannier localization constructs states \(|\widetilde\Phi_r^p(0)\rangle\) localized at holon–doublon separation \(r\), after which one computes the effective Hamiltonian matrix
\[
(\widetilde h^{\,p})_{r',r}
=\langle\widetilde\Phi_{r'}^p(0)|H_0^{(\mathrm C)}+H_V^{(\mathrm C)}-E_{gV}^{(\mathrm C)}|\widetilde\Phi_r^p(0)\rangle
\]
and solves
\[
\widetilde h_{\rm ex}|\psi_\lambda\rangle=\omega_\lambda|\psi_\lambda\rangle.
\]
This construction provides both qualitatively accurate spectra and direct real-space insight into weakly bound excitons in ET-F\(_2\)TCNQ [2006.09004].

For isolated atomic chains, the dominant ab initio workflow is DFT \(\rightarrow\) GW \(\rightarrow\) BSE. The cited study combines DFT(PBE), DFPT, single-shot \(G_0W_0\), eigenvalue-only self-consistent evGW, and BSE, with a cylindrical Coulomb cutoff to avoid image-charge interactions. Optical absorption for light polarized along the wire axis is expressed as
\[
A(\omega)=\frac{\omega}{c}\,\mathrm{Im}[\alpha_{1D}(\omega)],
\]
with \(\alpha_{1D}=(\varepsilon(\omega)-1)S/(4\pi)\) rendered independent of the transverse cell area \(S\). Within this framework, the variational soft-core and modified soft-core one-dimensional models reproduce the same chemical trends and come within \(\sim20\%\) of the ab initio binding energies, while underestimating exciton radii by roughly a factor of two [2510.09194].

A distinct development is the maximally-localized exciton Wannier-function formalism. Starting from exciton Bloch states \(\Psi_{S\mathbf Q}(\mathbf R,\mathbf r)\), a unitary gauge \(U_{SM}(\mathbf Q)\) generates localized exciton Wannier functions \(W_{M\mathbf R_\ell}(\mathbf R,\mathbf r)\) by minimizing the spread
\[
\Omega[U]=\sum_M\left[\langle R^2\rangle_M-\langle R\rangle_M^2\right].
\]
In quasi-1D materials, only the crystal momentum along the chain is sampled, and the 1D gauge optimization minimizes
\[
\Omega=\sum_M[\langle x^2\rangle_M-\langle x\rangle_M^2].
\]
The resulting representation supports exciton tight-binding Hamiltonians, Wannier–Fourier interpolation of \(E_S(Q_x)\), Zak-phase calculations, and interpolation of the exciton–phonon vertex [2308.03012]. Although the formalism was introduced in a three-dimensional bulk context, its explicit quasi-1D adaptation suggests a route toward compact effective Hamiltonians for chain and nanotube excitons.

At the opposite end of the complexity spectrum, a single-parameter correlated wavefunction remains useful for long nanorods:
\[
\Psi(\mathbf r_e,\mathbf r_h)=N\,\phi_e(\mathbf r_e)\phi_h(\mathbf r_h)e^{-a|x_e-x_h|}.
\]
Here the Slater-type factor \(e^{-a|x_e-x_h|}\) builds in the correct cusp at \(x_e=x_h\), the correlation length is \(\sim1/a\), and the long-axis kinetic energy acquires the closed form
\[
\langle T_x\rangle=\frac{\hbar^2}{2\mu_x}(k^2+a^2).
\]
This model captures the crossover from strong confinement to the quasi-1D Wannier-exciton regime while preserving compact analytical control over normalization, density, kinetic energy, and the Coulomb integral [1705.03232].

Taken together, these approaches indicate that quasi-one-dimensional Wannier excitons are best understood as a family of reduced-dimensional correlated states rather than a single model system. The same label encompasses weakly bound many-body excitons in Mott chains, extraordinarily robust trions in phosphorene, strongly bound yet Wannier-Mott–like excitons in isolated one-dimensional wires, and field-ionized resonances evolving into Wannier–Stark ladders. The unifying elements are one-dimensional relative motion, weak screening, and a real-space description in which localization and delocalization must be assessed simultaneously.

Source: https://www.emergentmind.com/topics/quasi-one-dimensional-wannier-excitons