Phonon-mediated electron attraction in SrTiO 3 via the generalized Fröhlich and deformation potential mechanisms

Christopher J. N. Coveney, Norm M. Tubman, Chih-En Hsu, Andres Montoya-Castillo, Marina R. Filip, Jeffrey B. Neaton, Zhenglu Li, Vojtech Vlcek, Antonios M. Alvertis · 2025

Read Phonon-mediated electron attraction in SrTiO 3 via the generalized Fröhlich and deformation potential mechanisms by Christopher J. N. Coveney, Norm M. Tubman, Chih-En Hsu, Andres Montoya-Castillo, Marina R. Filip, Jeffrey B. Neaton, Zhenglu Li, Vojtech Vlcek, Antonios M. Alvertis (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-17230.txt.

Introduction

Christopher Coveney and eight co-authors show how waves in a crystal's grid can pull electrons to pair, in a famed oxide that turns to a superconductor when doped. Their paper is from two thousand twenty-five.

That oxide was the first doped semiconductor found to superconduct, long ago, but the cause of its pairing is still fought over. The team builds a frame that lets the grid screen the push between electrons as it moves. They show, in clean math, that a broad form of the grid link can yield pairing here. First-principles counts then bare a next pull of the same kind. The work gives the old puzzle a firm new base.

Catalog record

CREATORChristopher J. N. Coveney, Norm M. Tubman, Chih-En Hsu, Andres Montoya-Castillo, Marina R. Filip, Jeffrey B. Neaton, Zhenglu Li, Vojtech Vlcek, Antonios M. Alvertis
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE33 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2501-17230

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Phonon-mediated electron attraction in SrTiO 3 via the generalized Fröhlich and deformation potential mechanisms Christopher J. N. Coveney, Norm M. Tubman, Chih-En Hsu, Andres Montoya-Castillo, Marina R. Filip, Jeffrey B. Neaton, Zhenglu Li, Vojtech Vlcek, Antonios M. Alvertis Abstract Superconductivity in doped SrTiO 3 was discovered in 1964, the first superconducting transition observed in a doped semiconductor. However, the mechanism of electron pairing in SrTiO 3 remains a subject of debate. By developing a theoretical framework to incorporate dynamical lattice screening in the electronic Coulomb interactions of semiconductors and insulators, we demonstrate analytically that linear long-range coupling of electrons to multiple longitudinal optical phonons, described by a generalized Fröhlich mechanism, can result in superconductivity in SrTiO 3 . Moreover, by combining our theory with first-principles calculations, we reveal an additional attractive interaction between electrons in SrTiO 3 due to the deformation potential mechanism, arising from the mixed ionic-covalent character of the Ti-O bond. Our results may have implications for the emergence of phonon-mediated electron attraction and superconductivity in a broad range of materials. The prediction that doped semiconductors can exhibit phonon-mediated superconductivity was made in 1964 in the seminal works by Cohen [ 1 , 2 ] , leading to the discovery of superconductivity in doped \text{SrTiO}_{3} in the same year [ 3 ] . Since then, superconductivity has been discovered in several semiconductors, including GeTe [ 4 , 5 ] , SnTe [ 6 , 7 ] , and MoS 2 [ 8 ] . Moreover, interfaces of \text{SrTiO}_{3} with LaAlO 3 and LaTiO 3 exhibit superconductivity [ 9 , 10 , 11 ] , generating interest in interface engineering to design high-temperature superconductors. Therefore, understanding the mechanism of \text{SrTiO}_{3} superconductivity is key towards designing new types of superconductors. To date, the precise mechanism of the pairing between electrons in \text{SrTiO}_{3} remains elusive. As \text{SrTiO}_{3} exhibits strong electron-phonon coupling [ 12 , 13 ] , several works have investigated the possibility of phonon-mediated superconductivity [ 1 , 14 , 15 , 16 , 17 , 18 , 19 ] . Proposed mechanisms include inter-valley scattering [ 1 , 14 ] , coupling to polar [ 15 ] , and defect [ 16 ] phonons, higher-order electron-phonon interactions [ 17 ] , coupling to the soft infrared active zone-center phonon of \text{SrTiO}_{3} [ 18 ] , which is enhanced by anharmonic damping [ 20 ] , and Rashba coupling [ 19 , 21 ] . However, the emergence of superconductivity through these mechanisms generally depends on parameters that are challenging to compute from first principles or to accurately infer from experiment [ 22 , 23 ] . A fully first-principles explanation for the phonon-mediated electron-electron attraction remains elusive, as \text{SrTiO}_{3} becomes superconducting at low doping densities where the Fermi energy is much smaller than the phonon frequencies, placing it in the non-adiabatic limit. As a result, the widely used Migdal-Eliashberg framework for phonon-mediated superconductivity does not apply [ 24 ] , leading to an ongoing debate on the mechanism of Cooper pair formation. Here we present a theory that rigorously incorporates lattice screening in electron-electron interactions. Without relying on adjustable parameters, we analytically demonstrate that lattice screening from multiple longitudinal optical (LO) phonons yields long-range attraction and superconductivity in \text{SrTiO}_{3} , in good agreement with experiments. Moreover, we translate our theory to a first-principles computational scheme, and show that the deformation potential associated with vibrations of the Ti-O bond provides an additional short-range attractive contribution to electron-electron interactions. Our theory, outlined in detail in our companion paper [ 25 ] , applies to several doped semiconductors and insulators. The density-density interaction between electronic states i,j is \displaystyle U_{i\mathbf{R}j\mathbf{R}^{\prime}}(\omega)= \displaystyle\int_{V}d\mathbf{r}\int_{V}d\mathbf{r}^{\prime}\phi_{i\mathbf{R}}^{*}(\mathbf{r})\phi_{i\mathbf{R}}(\mathbf{r})W(\mathbf{r},\mathbf{r}^{\prime},\omega)\phi_{j\mathbf{R}^{\prime}}^{*}(\mathbf{r}^{\prime})\phi_{j\mathbf{R}^{\prime}}(\mathbf{r}^{\prime}). (1) Here \phi_{i\mathbf{R}}(\mathbf{r}) are a basis localized in real space, specifically maximally localized Wannier functions (MLWFs) [ 26 ] centered around lattice vectors \mathbf{R},\mathbf{R}^{\prime} . Additionally, W(\omega) is the frequency-dependent screened Coulomb interaction, which has two contributions: an electronically screened Coulomb interaction W^{el} , and a contribution from phonons. Within many-body perturbation theory and to the lowest order in the electron-phonon interaction, the phonon contribution to the screened Coulomb interaction is written as [ 27 , 28 , 29 ] W^{ph}(\mathbf{r},\mathbf{r}^{\prime},\omega)=\sum_{\mathbf{q},\nu}D_{\mathbf{q},\nu}(\omega)g_{\mathbf{q},\nu}(\mathbf{r})g^{*}_{\mathbf{q},\nu}(\mathbf{r}^{\prime}). (2) Here D_{\mathbf{q},\nu}(\omega) is the propagator of a phonon with branch index \nu at wavevector \mathbf{q} , and g_{\mathbf{q},\nu} is the electron-phonon vertex. By inserting the overall screened Coulomb interaction W(\omega)=W^{el}(\omega)+W^{ph}(\omega) into eq. ( 1 ), we obtain the total interaction for electrons in states \phi_{i\mathbf{R}},\phi_{j\mathbf{R}^{\prime}} as U_{i\mathbf{R}j\mathbf{R}^{\prime}}(\omega)=U^{el}_{i\mathbf{R}j\mathbf{R}^{\prime}}(\omega)+U^{ph}_{i\mathbf{R}j\mathbf{R}^{\prime}}(\omega) . For the phonon-mediated interaction, as detailed in our companion paper [ 25 ] , we obtain for periodic systems \displaystyle U^{ph}_{i\mathbf{R}j\mathbf{R}^{\prime}}(\omega)=\bra{\phi_{i\mathbf{R}}\phi_{j\mathbf{R}'}}W^{ph}(\omega)\ket{\phi_{i\mathbf{R}}\phi_{j\mathbf{R}'}} \displaystyle=\sum_{\mathbf{q},\nu}g_{ii\mathbf{q}\nu}(\mathbf{0})g^{*}_{jj\mathbf{q}\nu}(\mathbf{R}^{\prime}-\mathbf{R}) \displaystyle\times[\frac{1}{\omega-\omega_{\mathbf{q},\nu}+i\delta}-\frac{1}{\omega+\omega_{\mathbf{q},\nu}-i\delta}], (3) where g_{mn\mathbf{q}\nu}(\mathbf{R})=\bra{\phi_{m\mathbf{R}}}g_{\mathbf{q}\nu}\ket{\phi_{n\mathbf{R}}} . Here we ignore non-local couplings of the form \bra{\phi_{m\mathbf{R}}}g_{\mathbf{q}\nu}\ket{\phi_{n\mathbf{R}'}} , which are expected to be small in the Wannier representation. In the static limit, and for on-site ( \mathbf{R}=\mathbf{R}^{\prime} ) intra-band ( i=j ) terms this becomes U^{ph,st}_{i\mathbf{R}i\mathbf{R}}=-2\sum_{\mathbf{q}\nu}\frac{|g_{ii\mathbf{q\nu}}(\mathbf{0})|^{2}}{\omega_{\mathbf{q},\nu}}, (4) which is a purely attractive interaction, in agreement with the conventional understanding of the effect of phonon screening on electron-electron interactions [ 30 ] . The superscript “ st ” denotes the static limit \omega=0 . In polar semiconductors and insulators, linear long-range dipolar electron-phonon coupling dominates the electron-phonon interaction [ 31 ] , and is written as \displaystyle g^{\mathcal{L}}_{ij\mathbf{q}\nu}(\mathbf{R})=i\frac{4\pi}{V}\sum_{\kappa}(\frac{1}{2NM_{\kappa}\omega_{\mathbf{q},\nu}})^{1/2}\frac{\mathbf{q}\cdot\mathbf{Z}_{\kappa}\cdot\mathbf{e}_{\kappa\nu}}{\mathbf{q}\cdot\mathbf{\epsilon}_{\infty}\cdot\mathbf{q}} \displaystyle\times\bra{\phi_{i\mathbf{R}}(\mathbf{r})}e^{i\mathbf{q}\cdot\mathbf{r}}\ket{\phi_{j\mathbf{R}}(\mathbf{r})}, (5) which is the generalization of the Fröhlich model [ 32 ] . Here we ignore local field effects, i.e. we set \mathbf{q}+\mathbf{G}\rightarrow\mathbf{q} , where \mathbf{G} a reciprocal lattice vector. For an atom \kappa , M_{\kappa} and \mathbf{Z}_{\kappa} are its mass and Born effective charge tensor respectively, \mathbf{e}_{\kappa\nu} a vibrational eigenvector, and \mathbf{\epsilon}_{\infty} the dielectric matrix tensor. Figure 1: Lowest conduction band manifold of \text{SrTiO}_{3} computed at the DFT-LDA level (black) and using Wannier interpolation (red) in panel a (energies relative to valence band maximum). The inset visualizes the unit cell of the system (Sr: green, Ti: gray, O: red). Panel b shows the phonon dispersion as obtained within DFPT including anharmonic corrections, alongside a visualization of the displacement patterns of the two highest frequency LO phonons. \epsilon_{\infty} \epsilon_{0} Z^{*}_{\text{Sr}} Z^{*}_{\text{Ti}} Z^{1*}_{\text{O}} Z^{2*}_{\text{O}} \omega_{TO-1},\omega_{LO-1} (meV) \omega_{TO-2},\omega_{LO-2} (meV) \omega_{TO-3},\omega_{LO-3} (meV) 6.2 409 2.56 7.25 -5.69 -2.05 65,94 21,55 11,21 Table 1: DFPT results for SrTiO 3 : high-and low-frequency dielectric constants \epsilon_{\infty},\epsilon_{0} , Born effective charges of different atoms (along two directions for oxygen), and frequencies of the pairs of TO,LO modes at \Gamma , for the three IR active LO phonons. By using the expression of eq. ( 5 ) in the phonon-mediated electron-electron interaction of eq. ( 3 ), taking the static limit, and assuming an isotropic dielectric and dispersionless phonons, we obtain the expression \displaystyle U^{ph,st,\mathcal{L}}_{\mathbf{R}\mathbf{R}^{\prime}}=-\frac{\gamma^{2}}{\epsilon_{\infty}|\mathbf{R}-\mathbf{R}^{\prime}|}, (6) where \gamma^{2}=\sum_{m}\frac{\omega_{LO,m}^{2}-\omega_{TO,m}^{2}}{\omega_{LO,m}^{2}} , and m runs over LO-TO phonon pairs of the system. This is an attractive interaction, with the contribution of each LO mode being proportional to the splitting \omega_{LO,m}^{2}-\omega_{TO,m}^{2} from the corresponding transverse optical (TO) mode. This result is derived within our companion paper [ 25 ] , alongside expressions for the general frequency-dependent case, which accounts for anisotropic dielectric and dispersive phonons. By comparing this expression to a repulsive electron-electron interaction 1/(\epsilon_{\infty}|\mathbf{R}-\mathbf{R}^{\prime}|) , it is evident that an overall long-range attraction appears when \gamma^{2}>1 . Superconductivity in \text{SrTiO}_{3} emerges upon doping of the conduction bands, it is therefore critical to incorporate doped

Read the complete text