Pith. sign in

REVIEW 3 major objections 4 minor 43 references

All stacking faults in both phases of GaN act as type II interfaces that locally reduce the band gap.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-03 10:05 UTC pith:ON2ZH44L

load-bearing objection First DFT study of zb-GaN stacking faults with a plausible but under-supported claim that all SFs are type II. the 3 major comments →

arxiv 2601.11206 v3 pith:ON2ZH44L submitted 2026-01-16 cond-mat.mtrl-sci

DFT modelling of stacking faults in hexagonal and cubic GaN

Zijie Wang , Mazharul M. Islam , David R. Bowler This is my paper
classification cond-mat.mtrl-sci
keywords stacking faultsGaNwurtzitezincblendeband offsetstype II band alignmentdensity functional theoryelectronic structure
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper uses density functional theory to examine every type of stacking fault—three intrinsic and one extrinsic in hexagonal wurtzite GaN, and intrinsic, extrinsic, and twin faults in cubic zincblende GaN—and asks what these faults do to the material's electronic structure. It finds that in every case the fault acts as a type II interface: the valence band maximum and conduction band minimum sit on opposite sides of the fault, and the local band gap is reduced relative to the perfect crystal. This directly contradicts the standard picture of a stacking fault as a thin inclusion of the other phase, which would predict the gap to rise in cubic GaN. The result matters because stacking faults are abundant in nitride devices, and their band alignment controls whether they trap carriers and shift the wavelengths of emitted light.

Core claim

The authors show that all seven stacking faults considered form type II band alignments: at each fault, both the valence band maximum and conduction band minimum rise to higher energies than in the bulk, and they localize on opposite sides of the fault plane. The local band gap is reduced by roughly 0.01 to 0.1 eV, with zincblende faults showing the larger reductions. This behavior is traced to a charge redistribution across the fault that creates an electrostatic potential step, which in wurtzite is opposite in sign to that in zincblende. The finding overturns the common 'inclusion' model of stacking faults in zincblende GaN as wurtzite insertions, since a wurtzite inclusion would locally r

What carries the argument

The central tool is the comparison of band edges between faulted supercells and perfect bulk using a common energy reference: the upper group of semi-core Ga d-bands, which has less than 0.1 eV dispersion, is aligned at the gamma point to set the zero. With this reference, the paper computes conduction and valence band offsets for each fault and classifies the alignment as type I or type II. The mechanism behind the offsets is established using band-resolved charge densities and the planar-averaged electrostatic potential difference between faulted and perfect cells, which shows how charge transfer across the fault creates the potential step that separates the band extrema spatially.

Load-bearing premise

The band-offset and gap-reduction conclusions assume that the upper group of semi-core Ga d-bands is an exact, transferable energy reference between a cell containing a stacking fault and a perfect bulk cell; if that reference shifts at the fault by just a few meV, the smallest reported offset (0.005 eV for the I1 fault) and the type II assignment for that fault would not hold.

What would settle it

Repeat the band-offset calculation for the I1 fault using a hybrid functional or GW, or align the supercell bands using core-level energies or the macroscopic average electrostatic potential far from the fault, and check whether the CBM offset stays at +5 meV; alternatively, map the local band gap across an isolated I1 fault with sub-10 meV energy resolution using low-temperature scanning tunneling spectroscopy.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • I1 is confirmed as the most stable stacking fault in wurtzite GaN, with a formation energy near 17 mJ/m², matching the experimental dominance of I1 faults in epilayers.
  • All three zincblende stacking faults have negative formation energies close to –33 mJ/m², so faults should form readily and with comparable abundance in cubic GaN.
  • Because each fault locally narrows the gap and separates electrons from holes, stacking faults are expected to act as low-energy recombination centers, consistent with cathodoluminescence and photoluminescence lines seen below the bulk band edge.
  • The type II classification rules out the common 'phase-inclusion' model of faults in optoelectronic device simulations; models that treat a fault as a quantum well of the other polymorph would misplace the band edges.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If this type II picture is confirmed by higher-level electronic-structure methods, stacking faults could be viewed as pre-existing charge-separation elements in GaN, potentially useful in photovoltaic or photocatalytic designs rather than only as harmful defects.
  • The I1 fault's conduction-band offset of about +5 meV is smaller than the quoted ≈0.1 eV dispersion of the d-band reference, so the type II classification of the most common wurtzite fault is the least numerically robust result in the paper.
  • The same charge-redistribution-and-potential-step mechanism should apply to other polymorphic III-V and II-VI semiconductors (e.g., AlN, InN, ZnS), suggesting a general rule that stacking faults in tetrahedrally bonded crystals introduce type II band offsets.
  • Since the three zincblende fault types are nearly degenerate in energy, any observed preference among them in growth comes from kinetics or local strain, not from thermodynamic stability, which could guide interpretation of transmission-electron-microscopy surveys.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 4 minor

Summary. The paper reports PBE-DFT calculations of stacking faults in wurtzite and zincblende GaN using the Conquest code with a PAO DZP basis. It considers all intrinsic, extrinsic, and twin fault geometries, computes formation energies, interplanar spacings, band-resolved charge densities, electrostatic potentials, projected DOS, and band offsets. The main claim is that all stacking faults in both phases form type II band alignments with a locally reduced band gap, contradicting the simple model of SFs as inclusions of the other phase. The formation-energy ordering for wurtzite (I1 < I2 < I3 < E) agrees with experiment, while all zincblende faults have similar, slightly negative formation energies. The paper includes convergence tests for the formation energies and makes the data openly available.

Significance. If the electronic-structure conclusions are robust, the paper would provide a systematic reference for how stacking faults modify the band edges of GaN, with implications for device modeling and for interpreting optical spectra. The study covers all fault types in both polytypes, which is a useful contribution. The formation-energy calculations are carefully converged with respect to cell size and match the known experimental preference for the I1 fault. However, the central band-offset and gap-reduction results rest on energy differences of 5–40 meV, and the paper currently provides no uncertainty estimates or validation of the common-reference scheme used for those offsets. The open data and detailed supporting information are strengths, but the quantitative electronic-structure claims are not yet secured to the precision required.

major comments (3)
  1. [Sec. II and Table VI (Appendix F)] The common energy zero is the Γ-point value of the upper semi-core Ga d-band group, justified by its <0.1 eV dispersion. However, the decisive I1 wz offsets in Table VI are ΔE_C = +0.005 eV and ΔE_g = −0.015 eV, an order of magnitude smaller than the reference dispersion. No test is reported showing that the d-band position relative to the VBM (or to the macroscopic average potential) is unchanged by the presence of the fault. Furthermore, the same table quotes Ref. [16] as giving ΔE_C = −0.045 eV for the same I1 fault, opposite in sign. Since the sign of ΔE_C is what distinguishes type II from type I, the central electronic-structure conclusion is not secured against a few-meV shift of the reference. The authors should validate the reference transferability (e.g., by computing offsets with macroscopic-average alignment and/or for multiple cell sizes) and report an uncertainty estimate.
  2. [Sec. IV and Table VI] The band offsets are reported without any convergence study with respect to supercell size, k-point grid, or basis-set cutoff. Given the offsets are 5–100 meV and the local gap changes are 10–40 meV, numerical noise at this scale could change the classification of the I1 fault. Please add a table or figure showing the dependence of ΔE_C, ΔE_V, and ΔE_g on the number of layers between faults (the data for Table V already exist, so this is a straightforward addition), and state the estimated numerical error.
  3. [Abstract vs. Sec. IV] The abstract claims all SFs 'can all be described as type II interfaces', but the text for the I1 wz fault says 'the conduction band minimum is extremely close to the energy in the bulk; the behavior is likely to be very sensitive to local fields and perturbations', and the conclusions repeat that the most common wz SF is 'highly sensitive to local conditions'. A 5 meV offset with a 0.1 eV reference dispersion cannot support a universal type II classification. Either the claim should be restricted to the non-I1 faults, or a quantitative confidence interval should be given that justifies including I1.
minor comments (4)
  1. [Table IV] Several c-axis lengths appear inconsistent; for example, 4AB-C-4AB lists c = 84.936 Å while the smaller 5AB-C-5AB cell lists 55.521 Å, and 7AB-14CB-7AB lists 279.823 Å, more than double the preceding value. Please check and correct the table.
  2. [Sec. IV] The phrase 'deep-lying non-dispersive d-bands' is imprecise, since Sec. II reports a dispersion of less than 0.1 eV for the upper group; please use the same wording consistently.
  3. [Sec. IV (discussion of experiment)] The experimental PL/CL shifts for SFs in wz (0.06–0.18 eV) are an order of magnitude larger than the calculated ΔE_g for I1 (−0.015 eV); a brief comment on the origin of the difference (e.g., excitonic or quantum-confinement effects) would help the reader.
  4. [Abstract] There is a typo: 'reducingthe gap' should be 'reducing the gap'.

Circularity Check

0 steps flagged

No circular derivation: DFT outputs are computed directly; the d-band alignment is an assumption, not a fitted input.

full rationale

The paper's load-bearing quantities—formation energies, band-resolved densities, potential differences, and band offsets—are direct outputs of periodic DFT calculations. Formation energies are obtained from Eq. (1) as total-energy differences (E_SF − n E_Ref), and the band offsets in Table VI are computed by comparing faulted and perfect cells after aligning the semi-core Ga d-bands. The choice of the 'upper group' of d-bands as a common zero is explicitly stated in Sec. II: 'This upper group forms a suitable reference point to compare the energies of different sets of bands, and we choose the value at the gamma point to set the common zero.' This is an open computational assumption justified by the small (<0.1 eV) dispersion of those bands; it is not a parameter fitted to reproduce the reported offsets. No fitted input is renamed as a prediction, and no uniqueness or ansatz result is imported from the authors' prior work to force the type-II classification. The self-citations in the paper are to the Conquest DFT code and PAO basis methodology ([22–24,28]); they are technical tool citations rather than load-bearing scientific claims. The d-band reference could affect the numerical robustness of the type-II conclusion (the I1 CBM offset is +0.005 eV, smaller than the quoted d-band dispersion) and the values disagree in sign with Ref. [16], but these are correctness/uncertainty concerns, not circularity: the claimed results are not equivalent to their inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 3 axioms · 0 invented entities

The paper's central claims rest on standard DFT approximations and on the choice of supercell and reference-level alignment; no free parameters are fitted to data. The main burden is the semi-core d-band common-zero assumption and the convergence of the zb supercells.

axioms (3)
  • domain assumption PBE-GGA exchange-correlation functional provides sufficiently accurate relative energetics and band-edge positions for GaN stacking faults.
    Sec. II: 'calculations used the generalized gradient approximation (GGA) based Perdew–Burke–Ernzerhof (PBE) exchange-correlation functional.' PBE is standard, but the paper acknowledges a large band-gap underestimate (1.65 eV vs 3.51 eV in wz), so the small type-II offsets could be functional-dependent.
  • domain assumption The upper group of semi-core Ga d-bands is a valid common energy reference between faulted and bulk cells.
    Sec. II: 'this upper group forms a suitable reference point to compare the energies of different sets of bands, and we choose the value at the gamma point to set the common zero.' The validity of this reference at faulted cells is not demonstrated, and its dispersion (~0.1 eV) is comparable to several quoted offsets.
  • ad hoc to paper The periodic supercells used in Table I are large enough that stacking-fault interactions do not affect the converged formation energies.
    Sec. III claims convergence testing ('to find the converged value of the formation energy'), but Table V shows non-monotonic variation for zb SFs (e.g., extrinsic ranges -31.2 to -36.3 mJ/m^2), so this assumption is not cleanly satisfied.

pith-pipeline@v1.3.0-alltime-deepseek · 11 in / 10811 out tokens · 228283 ms · 2026-08-03T10:05:20.508667+00:00 · methodology

0 comments
read the original abstract

We have performed density functional theory (DFT) calculations to characterize the energetics, and the atomic and electronic structure, of stacking faults in GaN, both in the stable hexagonal wurtzite (wz) phase and in the metastable cubic zincblende (zb) phase. In wz GaN, SFs on the (0001) planes can be divided into three different intrinsic stacking faults (I1, I2, and I3) and oneextrinsic stacking fault (E). In zb GaN, SFs form along (111) directions, giving one type each of intrinsic, extrinsic and twin SFs. Based on the calculated formation energy, I1 is the most stable SF of wz GaN in agreement with experiment. For zb GaN, the intrinsic stacking fault is the most dominant planar defect. To characterize the effect of the stacking faults on the electronic structure of the material, we examined the band density. We found that the bands near the valence band maximum in wz GaN are localised on the Ga-polar side of the stacking fault (i.e. on the Ga side of the Ga-N bonds perpendicular to the SF), with the bands near the conduction band minimum more on the N-polar side, though somewhat delocalised. We found the opposite trend in zb GaN; this behaviour is caused by a redistribution of charge near the interface. We also show the band offsets for the stacking faults, finding that they are very sensitive to local conditions, but can all be described as type II interfaces, with the presence of a stacking fault reducing the gap locally.

Figures

Figures reproduced from arXiv: 2601.11206 by David R. Bowler, Mazharul M. Islam, Zijie Wang.

Figure 2
Figure 2. Figure 2: FIG. 2: Stacking faults in Zincblende GaN. (a) Extrinsic [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 1
Figure 1. Figure 1: FIG. 1: Stacking faults in wurtzite GaN. (a) Intrinsic-1 [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 4
Figure 4. Figure 4: Band densities for all other SFs are given in the [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3: Band-resolved densities of Intrinsic-1 stacking [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4: Band-resolved densities of Intrinsic stacking [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6: Band edges: (a) WZ CBM; (b) ZB CBM; (c) [PITH_FULL_IMAGE:figures/full_fig_p006_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7: Local band gap plotted across simulation cell [PITH_FULL_IMAGE:figures/full_fig_p006_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8: Evolution of layer spacing along the c-axis of the simulation cell of the wz GaN SFs with respect to [PITH_FULL_IMAGE:figures/full_fig_p010_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: FIG. 9: Evolution of layer spacing along the c-axis of the simulation cell of the ZB GaN SFs with respect to [PITH_FULL_IMAGE:figures/full_fig_p011_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: FIG. 10: Band Density of Intrinsic-2 (4AB-8CA-C-4AB) stacking fault of wz GaN. (a) Band 591, (b) Band 592, (c) [PITH_FULL_IMAGE:figures/full_fig_p013_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: FIG. 11: Band Density of Intrinsic-3 (4AB-C-8BA-C-4AB) stacking fault of wz GaN. (a) Band 609, (b) Band 610, [PITH_FULL_IMAGE:figures/full_fig_p014_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: FIG. 12: Band Density of Extrinsic (6AB-C-6AB) stacking fault of wz GaN. (a) Band 447 , (b) Band 448 , (c) Band [PITH_FULL_IMAGE:figures/full_fig_p015_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: FIG. 13: Band Density of Extrinsic (4ABC-B-4ABC) stacking fault for zb GaN. (a) Band 447 , (b) Band 448 , (c) [PITH_FULL_IMAGE:figures/full_fig_p016_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: FIG. 14: Band density of Twin (4ABC-BAC-4ABC) stacking fault for zb GaN. (a) Band 483 , (b) Band 484 , (c) [PITH_FULL_IMAGE:figures/full_fig_p017_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: FIG. 15: Average potential of stacking faults based on wz GaN (a) I [PITH_FULL_IMAGE:figures/full_fig_p019_15.png] view at source ↗
Figure 16
Figure 16. Figure 16: FIG. 16: Average potential of stacking faults based on zb GaN (a) intrinsic (b) extrinsic and (c) twin. [PITH_FULL_IMAGE:figures/full_fig_p020_16.png] view at source ↗
Figure 17
Figure 17. Figure 17: FIG. 17: Projected density of states (pDOS) for wurtzite bulk and atoms located at the SFs, showing total (blue) [PITH_FULL_IMAGE:figures/full_fig_p022_17.png] view at source ↗
Figure 18
Figure 18. Figure 18: FIG. 18: Projected density of states (pDOS) for zincblende bulk and atoms located at the SFs, showing total (blue) [PITH_FULL_IMAGE:figures/full_fig_p023_18.png] view at source ↗
Figure 19
Figure 19. Figure 19: FIG. 19: Band gap evolution along c-axis of simulation cell for stacking faults in wz GaN. I [PITH_FULL_IMAGE:figures/full_fig_p025_19.png] view at source ↗
Figure 20
Figure 20. Figure 20: FIG. 20: Band gap evolution along c-axis of simulation cell for stacking faults of zb GaN. Intrinsic (a), Extrinsic (b) [PITH_FULL_IMAGE:figures/full_fig_p026_20.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

43 extracted references · 1 canonical work pages

  1. [1]

    Ding, Mater

    B. Ding, Mater. Sci. Technol.34, 1615 (2018)

    2018

  2. [2]

    T. Ito, T. Araki, T. Akiyama, and K. Nakamura, J. Crys. Growth301, 62 (2007)

    2007

  3. [3]

    Lester, F

    S. Lester, F. A. Ponce, M. G. Craford, and D. A. Steiger- wald, Appl. Phys. Lett.66, 1249 (1995)

    1995

  4. [4]

    Moram, C

    M. Moram, C. Ghedia, D. Rao, J. Barnard, Y. Zhang, M. Kappers, and C. Humphreys, Journal of Applied Physics106, 073513 (2009)

    2009

  5. [5]

    H¨ aberlen, T

    M. H¨ aberlen, T. J. Badcock, M. A. Moram, J. L. Hollan- der, M. J. Kappers, P. Dawson, C. J. Humphreys, and R. A. Oliver, J. Appl. Phys.108, 033523 (2010)

    2010

  6. [6]

    D. N. Zakharov, Z. Liliental-Weber, B. Wagner, Z. J. Reitmeier, E. A. Preble, and R. F. Davis, Phys. Rev. B 71, 235334 (2005)

    2005

  7. [7]

    W. Z. Tawfik, G. Y. Hyeon, and J. K. Lee, Journal of Applied Physics116, 164503 (2014)

    2014

  8. [8]

    Bernardini, V

    F. Bernardini, V. Fiorentini, and D. Vanderbilt, Phys. Rev. B56, R10024 (1997)

    1997

  9. [9]

    Morko¸ c, R

    H. Morko¸ c, R. Cingolani, W. Lambrecht, B. Gil, H.-X. Jiang, J. Lin, D. Pavlidis, and K. Shenai, MRS Internet J. Nitride Semic. Res.4, 18 (1999)

    1999

  10. [10]

    Trampert, O

    A. Trampert, O. Brandt, H. Yang, and K. Ploog, Appl. Phys. Lett.70, 583 (1997)

    1997

  11. [11]

    S. A. Church, M. Quinn, K. Cooley-Greene, B. Ding, A. Gundimeda, M. J. Kappers, M. Frentrup, D. J. Wallis, R. A. Oliver, and D. J. Binks, J. Appl. Phys.129, 175702 (2021)

    2021

  12. [12]

    Binks, P

    D. Binks, P. Dawson, R. Oliver, and D. Wallis, Appl. Phys. Lett.9, 041309 (2022)

    2022

  13. [13]

    B. Ding, M. Frentrup, S. Fairclough, M. Kappers, M. Jain, A. Kov´ acs, D. Wallis, and R. Oliver, J. Appl. Phys.128, 145703 (2020)

    2020

  14. [14]

    Stampfl and C

    C. Stampfl and C. G. Van de Walle, Phys. Rev. B59, 5521 (1999)

    1999

  15. [15]

    Batyrev, W

    I. Batyrev, W. Sarney, T. Zheleva, C. Nguyen, B. Rice, and K. Jones, phys. stat. sol. a208, 1566 (2011)

    2011

  16. [16]

    Benbedra, S

    A. Benbedra, S. Meskine, A. Boukortt, R. Hayn, M. Tex- ier, O. Thomas, and T. W. Cornelius, Comp. Cond. Mat- ter43, e01033 (2025)

    2025

  17. [17]

    Antoˇ s, P

    Z. Antoˇ s, P. Vacek, and R. Gr¨ oger, Comput. Mater. Sci. 180, 109620 (2020)

    2020

  18. [18]

    Barchuk, V

    M. Barchuk, V. Hol` y, D. Kriegner, J. Stangl, S. Schwaiger, and F. Scholz, Phys. Rev. B84, 094113 (2011)

    2011

  19. [19]

    Moram, C

    M. Moram, C. Johnston, J. Hollander, M. Kappers, and C. Humphreys, J. Appl. Phys.105(2009)

    2009

  20. [20]

    Y¨ ucelen, I

    E. Y¨ ucelen, I. Lazi´ c, and E. G. Bosch, Sci. Rep.8, 2676 (2018)

    2018

  21. [21]

    H. Xiu, S. M. Fairclough, A. Gundimeda, M. J. Kappers, D. J. Wallis, R. A. Oliver, and M. Frentrup, J. Appl. Phys.133, 105302 (2023)

    2023

  22. [22]

    D. R. Bowler, T. Miyazaki, and M. J. Gillan, J. Phys. Condens. Matter14, 2781 (2002)

    2002

  23. [23]

    Miyazaki and D

    T. Miyazaki and D. R. Bowler and R. Choudhury and M. J. Gillan, J. Chem. Phys.121, 6186 (2004)

    2004

  24. [24]

    Nakata, J

    A. Nakata, J. S. Baker, S. Y. Mujahed, J. T. L. Poulton, S. Arapan, J. Lin, Z. Raza, S. Yadav, L. Truflandier, T. Miyazaki, and D. R. Bowler,The Journal of Chemical Physics, J. Chem. Phys.152, 164112 (2020)

    2020

  25. [25]

    J. P. Perdew, K. Burke, and M. Ernzerhof, Phys. Rev. Lett.77, 3865 (1996)

    1996

  26. [26]

    D. R. Hamann, Phys. Rev. B88, 085117 (2013)

    2013

  27. [27]

    M. J. van Setten, M. Giantomassi, E. Bousquet, M. J. Verstraete, D. R. Hamann, X. Gonze, and G. M. Rig- nanese, Comp. Phys. Commun.226, 39 (2018)

    2018

  28. [28]

    D. R. Bowler, J. S. Baker, J. T. L. Poulton, S. Y. Muja- hed, J. Lin, S. Yadav, Z. Raza, and T. Miyazaki, Jap. J. Appl. Phys.58, 100503 (2019)

    2019

  29. [29]

    Schaefer, S

    B. Schaefer, S. Alireza Ghasemi, S. Roy, and S. Goedecker, J. Chem. Phys.142, 034112 (2015)

    2015

  30. [30]

    We will shorten this to AB and ABC respectively for simplicity of notation

    The stacking sequence of the layers in bulk GaN are often notated AaBb (wz) and AaBbCc (zb) with Aa indicating the alternation of planes of Ga and N along the (0001) di- rection (wz) or (111) direction (zb). We will shorten this to AB and ABC respectively for simplicity of notation

  31. [31]

    Ding,Study of wurtzite and zincblende GaN based green LED heterostructure, Ph.D

    B. Ding,Study of wurtzite and zincblende GaN based green LED heterostructure, Ph.D. thesis, University of Cambridge (2021)

    2021

  32. [32]

    Schulz and K

    H. Schulz and K. Thiemann, Solid State Commun.23, 815 (1977)

    1977

  33. [33]

    Frentrup, L

    M. Frentrup, L. Y. Lee, S.-L. Sahonta, M. J. Kappers, F. Massabuau, P. Gupta, R. A. Oliver, C. J. Humphreys, and D. J. Wallis, J. Phys. D: Appl. Phys.50, 433002 (2017)

    2017

  34. [34]

    Y. Gao, D. Sun, X. Jiang, and J. Zhao, J. Appl. Phys. 125, 215705 (2019)

    2019

  35. [35]

    Moram and M

    M. Moram and M. Vickers, Rep. Prog. Phys.72, 036502 (2009)

    2009

  36. [36]

    L¨ ahnemann, O

    J. L¨ ahnemann, O. Brandt, U. Jahn, C. Pf¨ uller, C. Roder, P. Dogan, F. Grosse, A. Belabbes, F. Bechstedt, A. Trampert, and L. Geelhaar, Phys. Rev. B86, 081302 (2012)

    2012

  37. [37]

    M. A. Caro, S. Schulz, and E. P. O’Reilly, Phys. Rev. B 88, 214103 (2013)

    2013

  38. [38]

    Vurgaftman and J

    I. Vurgaftman and J. R. Meyer, J. Appl. Phys.94, 3675 (2003)

    2003

  39. [39]

    X. H. Lu, P. Y. Yu, L. X. Zheng, S. J. Xu, M. H. Xie, and S. Y. Tong, Appl. Phys. Lett.82, 1033 (2003)

    2003

  40. [40]

    Y. J. Sun, O. Brandt, U. Jahn, T. Y. Liu, A. Trampert, S. Cronenberg, S. Dhar, and K. H. Ploog, J. Appl. Phys. 92, 5714 (2002)

    2002

  41. [41]

    R. Liu, A. Bell, F. A. Ponce, C. Q. Chen, J. W. Yang, and M. A. Khan, Applied Physics Letters86, 021908 (2005)

    2005

  42. [42]

    Z. Wang, M. Islam, and D. R. Bowler, https://doi.org/10.5281/zenodo.18259436 (2026)

    work page doi:10.5281/zenodo.18259436 2026

  43. [43]

    L. C. De Carvalho, A. Schleife, and F. Bechstedt, Phys. Rev. B84, 195105 (2011)

    2011