Spin-orbit interaction in square core-shell nanowires

Andrei Manolescu; Anna Sitek; Sigurdur I. Erlingsson; Tudor Gabriel Dumitru

REVIEW 3 major objections 6 minor 17 references

Spin-orbit interaction in square core-shell nanowires

T0 review · 3 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Spin-orbit interaction turns four corner states into independent quantum wires.

desk verdict A solid, modest k.p study of Rashba SOI in square core-shell nanowires; the qualitative picture of weakly coupled corner wires is believable, but the central claim needs a quantitative coupling estimate. read the letter →

arxiv 2506.02769 v1 pith:YACPAX7W submitted 2025-06-03 cond-mat.mes-hall

Anna Sitek , Tudor Gabriel Dumitru , Sigurdur I. Erlingsson , Andrei Manolescu This is my paper

classification cond-mat.mes-hall

keywords core-shellnanowiresspin-orbitinteractionRashbaeffectsquarecross-sectioncornerstatesquantumwiresk·pmethodband-offsetpotential

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper predicts that, in a square core-shell nanowire with an 8-nm InAs shell on an InP core, low-energy electrons sit in the four corners and are separated from side-localized states by a substantial gap. With Rashba spin-orbit interaction coming from the graded core-shell potential, the eight corner states organize into four pairs that almost do not interact, so a single wire mimics four independent quantum wires; the higher side states instead anti-cross like coupled wires. The authors derive this from a k·p Hamiltonian solved on a polar grid with square boundaries, treating both the extended band-offset potential and an external electric field. This matters because one physical nanowire could supply four nearly identical, weakly coupled one-dimensional channels, with coupling turning on at higher energies.

What carries the argument

The load-bearing object is the single-particle k·p Hamiltonian for conduction electrons in the shell, Eq. (1), whose Rashba coefficients are derivatives of the total confining potential, Eqs. (2). It is diagonalized on a polar grid whose points are masked to the square cross-section with Dirichlet boundary conditions, following a finite-difference method used for polygonal rings. The spin-orbit interaction is generated either by the extended band-offset potential, taken to drop linearly from 656.6 meV to zero across the full 8-nm shell so its gradient is non-zero everywhere, or by an external electric field that breaks all symmetries of the square. The folding-down procedure supplies this two-band conduction Hamiltonian from the coupling to valence bands, and the transverse states at kz=0 are grouped in sets of eight, twice the number of corners.

What would settle it

An atomistic tight-binding calculation of the same 8-nm InAs/InP square wire, or tunneling spectroscopy of its low-energy levels, that does not show the predicted corner-side gap near $\Delta \approx 19$ meV and four nearly flat corner bands over $k_z = \pm 8 \times 10^7$ m$^{-1}$ would falsify the four-independent-wires picture.

Watch

Extended reading notes

Core claim

The central claim is that spin-orbit interaction reshapes the eight lowest transverse states of a square core-shell nanowire into the spectrum of four parallel, nearly identical quantum wires. At kz=0 these states form one twofold, one fourfold, and one twofold degenerate level; once Rashba coupling from the extended band-offset potential is included, the fourfold degenerate branches (l=2 and l=3) move toward their neighbours and the four pairs track the four corners, crossing or barely anti-crossing along kz. By contrast, the first group of side-localized states (l=5 to l=8) shows much larger dispersion and splits into two pairs of closely spaced bands with significant anti-crossings, the fingerprint of coupled wires. An external electric field breaks the square symmetries, lifts the fourfold degeneracy, and shifts spin pairs to ±kz, yet the corner-side gap remains large enough to address the corner channels selectively.

Load-bearing premise

The quantitative band structure assumes the intermixing potential between core and shell falls linearly from 656.6 meV to zero over the full 8-nm shell, with electrons strictly confined by Dirichlet boundaries; a different intermixing profile or leakage of surface states would change the corner-side gap and the spin-orbit splittings.

Editorial extensions

If this is right

At low energies a single square core-shell nanowire can be used as four parallel, nearly identical one-dimensional quantum wires confined to the corners.
At higher energies the same structure switches to coupled-wire behaviour, with side-localized states showing significant anti-crossings and possible level-order interchange.
With the graded potential alone, spin-orbit interaction preserves twofold spin degeneracy at all kz and only lifts fourfold degeneracies; an external electric field is needed to produce ±kz-shifted spin-split branches.
The corner-side gap, though reduced by an electric field, stays large enough that corner states can be manipulated without populating side states.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If the corner-wire mechanism is generic, polygonal shells with N corners should show N independent low-energy channels; checking triangular or hexagonal InAs shells would show whether the square case is representative.
Because the Rashba splittings come from the graded intermixing potential, varying the intermixing length or profile during growth should continuously tune both the corner-side gap and the effective inter-wire coupling, giving an experimental control knob the paper does not quantify.
A transport experiment along the wire should see the low-energy channels as four parallel conducting paths: conductance steps in multiples of $\frac{2e^2}{h}$ (or $\frac{e^2}{h}$ per spin-resolved channel) might reveal the four-wire structure directly.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Quick take: this is a solid, modest k.p study of Rashba spin-orbit effects in square core-shell nanowires. The genuinely new content is the SOI-modified dispersion for the square geometry: corner states remain nearly degenerate and weakly coupled, while side states show clear anti-crossings. The qualitative picture is believable and the numerics look standard. The main soft spots are that the 'independent wires' claim is never quantified, and the model leans on an assumed linear intermixing profile without sensitivity analysis.

What's good: the paper connects the known corner localization in polygonal shells to SOI. The degeneracy structure at kz=0 is correctly explained. The figures are informative, and the distinction between corner and side states is convincing. Using a k.p method with a scalar Rashba term is standard, and the conclusion that low-energy corner states form quasi-1D channels is plausible and relevant for Majorana and spin-orbit qubit proposals.

Where it's soft: (1) No coupling matrix element, anti-crossing gap, or energy scale is given to support the independent-wires claim. The stress-test note is on point: 'nearly identical' is a visual judgment. If the corner manifold spread is a few meV while the SOI splitting is tens of micro-eV, the four channels are not truly independent. A quantitative measure would make the paper much stronger. (2) The intermixing potential is assumed linear from 656.6 meV to zero over the 8-nm shell. That's a reasonable guess, but it directly controls the SOI strength and the corner-side gap. No sensitivity analysis or comparison to more realistic profiles is provided. (3) The folding-down derivation is referenced rather than sketched, and key material parameters (m*, alpha0) are not given in the text. That hurts reproducibility. (4) The self-citation of refs 8, 10, 12 is fine, since corner localization is their earlier work.

These concerns are not dealbreakers. The qualitative conclusion doesn't hinge on exact numerical coupling: the paper's claim is 'nearly independent' and 'minimal anti-crossings,' which accurately describes the figures. The real gap is that a design-oriented reader can't tell if the independence survives thermal or experimental energy scales.

Who is this for? Researchers working on core-shell nanowires for spin-orbit qubits or Majorana devices. It's a useful reference for the square-geometry band structure, and it deserves peer review. A referee should ask for quantitative coupling estimates and better parameter documentation.

My recommendation: engage with it, but treat the independent-wires conclusion as a hypothesis to test quantitatively, not as a proven fact.

Referee Report

3 major / 6 minor

Summary. The manuscript studies Rashba spin-orbit interaction in electrons confined to the shell of a square InAs/InP core-shell nanowire. The authors diagonalize a two-band k.p Hamiltonian with a kz-linear Rashba term on a polar finite-difference grid with Dirichlet boundary conditions, using an extended band-offset potential that decreases linearly over the shell width. They first present the transverse states of a square quantum ring, showing groups of eight states with corner- and side-localized probability distributions. With spin-orbit interaction from the extended band-offset potential alone, and then with an additional external electric field, they compute subband dispersions. They interpret the low-energy corner-localized states as four nearly independent quantum wires, while higher-energy side-localized states are interpreted as coupled wires.

Significance. If the interpretation is established quantitatively, the paper would be useful for proposals using single core-shell nanowires as multichannel systems, for example for multiple Majorana states or selective manipulation of corner channels. The numerical method is standard, no parameter is fitted to data, and the qualitative localization picture and degeneracy structure are clearly presented. However, the central "independent wires" conclusion currently rests on visual near-degeneracy and the apparent size of anti-crossings, rather than on a quantitative coupling energy scale. A revision that supplies numerical gaps and a sensitivity analysis would make the claim convincing.

major comments (3)

[Section 3.2, Figs. 2 and 3] The claim that the eight lowest corner-localized states resemble independent quantum wires is not quantitatively supported. The text infers "minimal anti-crossings" from the plots, but it never gives the avoided-crossing gap or a hopping matrix element for the corner states, nor compares these with the SOI-induced spin splitting or with a relevant thermal or experimental energy scale. For four independent channels, the geometric splitting at kz=0 and the inter-corner tunneling matrix element must be much smaller than the SOI energy scale at the relevant kz. Please report numerical values, for example the minimum anti-crossing gaps for the pairs l=2/l=3 in Fig. 2 and n=4/n=5 in Fig. 3, together with the zero-field spread of the l=1..4 manifold; without these numbers, the central conclusion is an interpretation rather than a demonstrated result.
[Section 2 and Figs. 2-3] The quantitative conclusion depends on the assumed linear intermixing potential: VBO decreases linearly from 656.6 meV to zero over a radial distance r, with r taken equal to the 8 nm shell width, and Dirichlet boundary conditions are imposed at the square cross-section. This potential profile is not validated against atomistic simulation or experimental data, and it directly controls both the corner-side gap and the SOI-induced splittings in Figs. 2 and 3. Please add a sensitivity analysis, for example varying r or replacing the linear profile by a smooth or error-function profile, and test the choice of boundary conditions, to show that the independent-wire picture is robust.
[Fig. 3 caption and Section 3.2] The external electric field amplitude is never specified. Because the field strength controls the level crossings and anti-crossings that support the coupled-wire interpretation, the figure cannot be reproduced or checked without this parameter. State the field magnitude and its exact orientation in the coordinate frame used, either in the caption or in the text.

minor comments (6)

[Section 2, Eq. (1)] The notation "12×2" is ambiguous; please write the identity matrix as I2 and specify the matrix dimensions explicitly. Also define alpha0 and list the material parameters used (effective mass, gap, spin-split gap, interband matrix elements).
[Section 3.1, Fig. 1] The geometry of the square ring is not fully specified: "8-nm-wide InAs shell deposited on a 57-nm-thick InP core" does not uniquely determine the inner and outer side lengths of the square cross-section. Please give the exact dimensions used in the calculation.
[Section 3.2, Fig. 3(a)] The phrase "exhibits a much larger energy dispersion at kz=0" for the second group of corner-localized states is confusing; at kz=0 there is no dispersion. This appears to mean the energy spread of the group at kz=0. Please rephrase.
[Section 3.2, Fig. 2] The sentence "the states originating from the fourfold degenerate state at kz=0 (l=2 and l=3)" is unclear because at kz=0 the SOI term vanishes, so l=2 and l=3 are degenerate there and only at finite kz do they split and cross or anti-cross. Please clarify the wording.
[References] Reference [1] contains a typo: "GaGs" should be "GaAs".
[Methods, folding-down procedure] The folding-down procedure is cited to Refs. [16,17] but no derivation or explicit expression for alpha0 is given. For self-containedness, include the relevant formula or a brief derivation in the Methods section.

Circularity Check

0 steps flagged · score 1.0 of 10

No circular derivation: the model is solved from stated potentials and material parameters; the four-wire picture is an interpretation of the computed dispersions, not an input or fitted output.

full rationale

The calculation is self-contained: Eq. (1) is a k.p Hamiltonian with Rashba coefficients defined in Eqs. (2) as derivatives of the assumed extended band-offset potential, and the spectrum is obtained by direct diagonalization on a polar grid with Dirichlet boundary conditions. No parameter is fitted to the computed bands and no quantity from Figs. 2 or 3 is used as an input to the model. The central qualitative claim—that the eight lowest corner-localized states resemble four independent quantum wires—is a reading of the computed dispersions (near-degeneracy, simple crossings, and 'minimal anti-crossings' at finite kz) rather than a restatement of the Hamiltonian. The self-citations [8,10,12] establish the known corner-localization geometry and are used as background for choosing the square-shell model; they are not invoked to force the SOI result. The linear intermixing profile (656.6 meV dropping over 8 nm) is an unvalidated modeling assumption and feeds into the quantitative splittings, but this is a correctness/robustness concern, not circularity: the assumption does not already contain the conclusion that the states behave as four independent wires. A more quantitative support for the independence claim (e.g., explicit inter-corner hopping or avoided-crossing gap) would be desirable, but its absence is not a circular-step defect under the stated criteria.

Assumptions & free parameters 2 free parameters · 5 assumptions · 0 invented entities

The calculation does not introduce new particles or forces. It rests on standard k.p and Rashba models plus an assumed linear intermixing potential; the only hand-chosen free parameters are the intermixing length and the (unstated) electric-field amplitude, both of which set the quantitative band structure shown in Figs. 2 and 3.

free parameters (2)

Intermixing length r = 8 nm (chosen equal to shell width)
Controls the spatial extent of the band-offset potential and therefore the Rashba field; the paper argues the strongest SOI occurs when r spans the shell, and Fig. 2 uses r=8 nm. Not fitted to data, but a free model choice on which the quantitative dispersions depend.
External electric field amplitude = not reported
Fig. 3 applies a field rotated 30 degrees from the diagonal but the text never states the field strength, so the plotted dispersions cannot be reproduced from the paper.

assumptions (5)

domain assumption Conduction-band electrons in the shell are described by a k.p effective-mass Hamiltonian with a Rashba term (Eq. 1).
Invoked in Section 2; standard for narrow-gap semiconductors but neglects valence-band mixing beyond the folding-down procedure.
domain assumption Rashba coefficients are proportional to the gradient of the total potential (Eq. 2).
Standard Rashba model; assumes the spin-orbit field follows the local electric field.
ad hoc to paper The extended band-offset potential decreases linearly from 656.6 meV to zero over the radial distance r and is zero in the clean-shell limit.
Introduced in Section 2 to model core-shell intermixing; no atomistic or experimental validation of the linear profile.
domain assumption Electron wavefunctions vanish at the outer boundary of the nanowire (Dirichlet boundary conditions).
Mentioned in Section 2; ignores surface states or finite barrier outside the wire.
standard math A polar grid with polygonal boundaries faithfully represents the square cross-section with 8 nm shell and 57 nm core.
Numerical discretization choice; convergence or grid-spacing checks are not reported.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Spin-orbit interaction in square core-shell nanowires." pith.science (2026). https://pith.science/paper/YACPAX7W

@misc{pith2026250602769,
  author       = {Pith},
  title        = {Pith review of: Spin-orbit interaction in square core-shell nanowires},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YACPAX7W}},
  note         = {Machine review of arXiv:2506.02769}
}

read the original abstract

We theoretically investigate the spin-orbit interaction of electrons confined in the outer regions of square core-shell nanowires. The polygonal cross section leads to the accumulation of low-energy electrons in the corners and the formation of a significant energy gap that separates these corner-localized states from higher-energy states localized along the sides. We show that the low-energy states behave like the states of independent quantum wires, while the higher-energy states exhibit features characteristic of coupled wires.

Figures

Figures reproduced from arXiv: 2506.02769 by the authors.

**Figure 1.** (a) Low-energy transverse states (∆ = 19 meV, δ = 9 meV). (b)-(j) Electron probability distributions corresponding to the energy levels shown in panel (a). αx(x, y) = −α0 ∂ ∂y V (x, y) (2a) and αy(x, y) = α0 ∂ ∂xV (x, y) (2b) The amplitude α0 depends on the shell material semiconductor band gap, spin-split gap and interband momentum matrix elements. Finally, σx and σy represent the Pauli matrices. We diagonalize the… view at source ↗

**Figure 2.** Energy dispersions in the presence of SOI induced by the extended bandoffset potential. strongest SOI effect occurs when the contribution from core molecules gradually decreases throughout the entire shell, such that the extended band-offset potential reaches zero at the outer boundaries of the wire. In this case, the derivatives in Eqs. 2, and thus the Rashba coefficients are non-zero across the entire cross-sec… view at source ↗

**Figure 3.** (a) Energy dispersions in the presence of SOI due to the extended band-offset potential and the external electric field. (b)-(m) The corresponding probability distributions. 3 [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

17 extracted references · 17 canonical work pages

[1]

Jadczak, et al

J. Jadczak, et al. , Unintentional high-density p-type modulation doping of a G aGs/AlAs core–multishell nanowire, Nano Lett. 14, 2807 (2014)

work page 2014
[2]

Bl¨ omers,et al., Realization of nanoscaled tubular conductors by means of Ga As/InAs core/shell nanowires, Nanotechnology 24, 035203 (2013)

C. Bl¨ omers,et al., Realization of nanoscaled tubular conductors by means of Ga As/InAs core/shell nanowires, Nanotechnology 24, 035203 (2013)

work page 2013
[3]

M. M. Sonner, et al., Breakdown of corner states and carrier localization by mono layer ﬂuctuations in radial nanowire quantum wells, Nano Lett. 19, 3336 (2019)

work page 2019
[4]

L. F. Pe˜ na, et al., Modeling Si/SiGe quantum dot variability induced by interf ace disorder reconstructed from multiperspective microscopy, npj Quantum Inf 10, 33 (2024)

work page 2024
[5]

Qian, et al., Gallium nitride-based nanowire radial heterostructures for nanophotonics, Nano Lett

F. Qian, et al., Gallium nitride-based nanowire radial heterostructures for nanophotonics, Nano Lett. 4, 1975 (2004)

work page 2004
[6]

G¨ uniat,et al., III–I integration on Si(100): V ertical nanospades , ACS Nano 13, 5833 (2019)

L. G¨ uniat,et al., III–I integration on Si(100): V ertical nanospades , ACS Nano 13, 5833 (2019)

work page 2019
[7]

Aruni Fonseka, et al., Engineering the side facets of vertical [100] oriented InP n anowires for novel radial heterostructures, Nanoscale Research Lett

H. Aruni Fonseka, et al., Engineering the side facets of vertical [100] oriented InP n anowires for novel radial heterostructures, Nanoscale Research Lett. 14, 399 (2019)

work page 2019
[8]

Sitek, et al

A. Sitek, et al. , Electron localization and optical absorption of polygonal quantum rings , Phys. Rev. B 91, 235429 (2015)

work page 2015

Show all 17 references

[9]

Ballester, et al., Multi-particle states of semiconductor hexagonal rings: A rtiﬁcial benzene , J

A. Ballester, et al., Multi-particle states of semiconductor hexagonal rings: A rtiﬁcial benzene , J. Appl. Phys 112, 104317 (2012)

2012
[10]

Sitek, et al., Multi-domain electromagnetic absorption of triangular qu antum rings, Nanotechnology 27, 225202 (2016)

A. Sitek, et al., Multi-domain electromagnetic absorption of triangular qu antum rings, Nanotechnology 27, 225202 (2016)

2016
[11]

Ferrari, et al., Magnetic states in prismatic core multishell nanowires , Nano Lett

G. Ferrari, et al., Magnetic states in prismatic core multishell nanowires , Nano Lett. 9, 1631 (2009)

2009
[12]

Manolescu, et al., Majorana states in prismatic core-shell nanowires , Phys

A. Manolescu, et al., Majorana states in prismatic core-shell nanowires , Phys. Rev. B 96, 125435 (2017)

2017
[13]

T. D. Stanescu, et al. , Robust topological phase in proximitized core–shell nanow ires coupled to multiple superconductors, Beilstein J. Nanotechnol. 9, 1512 (2018)

2018
[14]

T. D. Stanescu, Introduction to T opological Quantum Matter and Quantum Computation, CRC Press, Oxford (2017)

2017
[15]

Nadj-Perge, et al., Spin–orbit qubit in a semiconductor nanowire , Nature 468, 1084 (2010)

S. Nadj-Perge, et al., Spin–orbit qubit in a semiconductor nanowire , Nature 468, 1084 (2010)

2010
[16]

Fabian, et al., Semiconductor spintronics, Acta Phys

J. Fabian, et al., Semiconductor spintronics, Acta Phys. Slovaca 57, 565 (2007)

2007
[17]

W´ ojcik,et al., Anisotropy of the spin-orbit coupling driven by a magnetic ﬁ eld in InAs nanowires , Phys

P . W´ ojcik,et al., Anisotropy of the spin-orbit coupling driven by a magnetic ﬁ eld in InAs nanowires , Phys. Rev. B 103, 085434 (2021). 4

2021

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

Jadczak, et al

J. Jadczak, et al. , Unintentional high-density p-type modulation doping of a G aGs/AlAs core–multishell nanowire, Nano Lett. 14, 2807 (2014)

work page 2014

[2] [2]

Bl¨ omers,et al., Realization of nanoscaled tubular conductors by means of Ga As/InAs core/shell nanowires, Nanotechnology 24, 035203 (2013)

C. Bl¨ omers,et al., Realization of nanoscaled tubular conductors by means of Ga As/InAs core/shell nanowires, Nanotechnology 24, 035203 (2013)

work page 2013

[3] [3]

M. M. Sonner, et al., Breakdown of corner states and carrier localization by mono layer ﬂuctuations in radial nanowire quantum wells, Nano Lett. 19, 3336 (2019)

work page 2019

[4] [4]

L. F. Pe˜ na, et al., Modeling Si/SiGe quantum dot variability induced by interf ace disorder reconstructed from multiperspective microscopy, npj Quantum Inf 10, 33 (2024)

work page 2024

[5] [5]

Qian, et al., Gallium nitride-based nanowire radial heterostructures for nanophotonics, Nano Lett

F. Qian, et al., Gallium nitride-based nanowire radial heterostructures for nanophotonics, Nano Lett. 4, 1975 (2004)

work page 2004

[6] [6]

G¨ uniat,et al., III–I integration on Si(100): V ertical nanospades , ACS Nano 13, 5833 (2019)

L. G¨ uniat,et al., III–I integration on Si(100): V ertical nanospades , ACS Nano 13, 5833 (2019)

work page 2019

[7] [7]

Aruni Fonseka, et al., Engineering the side facets of vertical [100] oriented InP n anowires for novel radial heterostructures, Nanoscale Research Lett

H. Aruni Fonseka, et al., Engineering the side facets of vertical [100] oriented InP n anowires for novel radial heterostructures, Nanoscale Research Lett. 14, 399 (2019)

work page 2019

[8] [8]

Sitek, et al

A. Sitek, et al. , Electron localization and optical absorption of polygonal quantum rings , Phys. Rev. B 91, 235429 (2015)

work page 2015