The 1977 analysis of Frensley and Kroemer extracted effective ionic charges in zinc-blende-type semiconductors from the difference of the crystal electrostatic potential between the octahedral and tetrahedral interstitial sites, finding charges proportional to Phillips electronegativity differences (e∗/e≅0.76 ΔXPhe^*/e \cong 0.76\,\Delta X_{Ph} for 13 compounds), and proposed the interstitial potential as a bulk reference for absolute semiconductor band energies. Their own follow-up paper completed the heterojunction dipole application, and modern first-principles calculations (2021) have reproduced and extended the dataset — yet two issues have remained open for five decades: (1) the interstitial-potential charges lack any quantitative microscopic reconciliation with charges derived from infrared (dynamic/Szigeti) measurements, LCAO theory, modern Bader density partitioning, and Phillips spectroscopic ionicity (the 2021 study compares the interstitial-potential charges with Bader partial charges and ionicity measures without providing such a reconciliation); and (2) there is no first-principles error theory for the interstitial-potential band reference — the consequences of the neglected interstitial charge density, the non-spherical bond charge around the tetrahedral site, and the ionic-model dipole corrections are documented (the method's accuracy is reported to degrade for compounds with small lattice constants) but are predicted by no theory. This problem asks for a quantitative microscopic account that reconciles the effective-charge measures across the compound family and a predictive error theory of the interstitial reference, addressing the documented discrepancies and accuracy limitations.

Background

Zinc-blende-type semiconductors (the crystal structure of ZnS, GaAs, and most III-V and II-VI compounds) consist of two interpenetrating face-centered-cubic sublattices, one for each chemical species, displaced by a quarter of the body diagonal. The structure contains two inequivalent high-symmetry empty positions — interstitial sites: the octahedral site at (1/2, 1/2, 1/2) and the tetrahedral site at (3/4, 3/4, 3/4), respectively the octahedral hole and an unoccupied tetrahedral hole of one fcc sublattice (the remaining tetrahedral holes of that sublattice are occupied by the second species' atoms). Because the two atomic species carry different charges, the electrostatic potential is not flat in the void regions, and the potential difference ΔV\Delta V between the two sites is a purely electrostatic, directly computable measure of the compound's polarity. In a point-ion model with effective charges on the two sublattices, ΔV\Delta V is fixed by Madelung-type lattice sums of the ionic Coulomb potentials; the relevant dimensionless constant follows from the observation that filling the empty interstices by superimposing a second, displaced zinc-blende lattice would produce the cesium-chloride structure, giving αCsCl−αZnS=0.1246\alpha_{CsCl} - \alpha_{ZnS} = 0.1246 (the Madelung constants of the two structure types, referred to the nearest-neighbor distance). Frensley and Kroemer found from empirical-pseudopotential band-structure calculations (the 1970s method in which the crystal potential is fitted to experimental band-structure data) that for 13 zinc-blende compounds ΔV=[Q(B)−Q(A)]/a\Delta V = [Q(B) - Q(A)]/a (with lattice constant aa) is reproduced by charges Q=6.27 eVA˚×XPhQ = 6.27\ \mathrm{eV\AA} \times X_{Ph} proportional to the Phillips electronegativity XPhX_{Ph} — a spectroscopic scale derived from dielectric data — equivalent to e∗/e=ΔQ/(8.287 eVA˚)≅0.76 ΔXPhe^*/e = \Delta Q/(8.287\ \mathrm{eV\AA}) \cong 0.76\,\Delta X_{Ph}.

The effective ionic charge of a partially covalent crystal is not a unique observable but a family of metrics, each defined by an operational prescription: (i) the static charge extracted from the electrostatic potential in the interstitial voids (the Frensley-Kroemer quantity); (ii) dynamic charges — the transverse (Szigeti) charge from infrared lattice-vibration spectroscopy and the Born effective charge from the polarization induced by displacing an atom — which can exceed the formal charge in covalent crystals; (iii) LCAO (linear-combination-of-atomic-orbitals) population charges from tight-binding theory; (iv) Bader charges, obtained by partitioning the computed electron density at zero-flux surfaces; and (v) the Phillips ionicity fif_i, a dielectric/spectroscopic measure on a 0-1 scale. That these metrics disagree — often substantially — for the same compound is a long-standing feature of the microscopic description of partially covalent solids: each is sensitive to a different physical quantity (static potential, dynamic response, or density partition), and no quantitative theory relates them.

A semiconductor heterojunction — the interface between two different semiconductors — has a valence-band offset that is the sum of the difference of the two materials' bulk valence-band energies relative to a common reference and a dipole correction from the redistribution of charge at the interface. The naive electron-affinity rule (band offset equals the difference of electron affinities, i.e., of bulk levels each referenced to the vacuum outside that material) neglects the interface dipole and fails in general. The Frensley-Kroemer program, developed in their 1976 paper and completed in the two 1977 papers, replaced the vacuum reference with a bulk internal reference — the interstitial electrostatic potential — whose difference between materials, added to band-structure terms, gives the lineup; the same ionic charges that reproduce ΔV\Delta V also predict the interface dipole. The related absolute band energies are the ionization potential (IP, the energy to remove an electron from the valence-band maximum, referenced to vacuum) and the electron affinity (EA).

The interstitial reference was revived as a rapid estimator of IP and EA that avoids the slab supercells otherwise needed to define a vacuum level. Modern density-functional theory (DFT) calculations with projector-augmented-wave (PAW) potentials and functionals from the local-density approximation (LDA) to the hybrid functional HSE06 reproduce the original dataset in qualitative terms (Harnett-Caulfield and Walsh, 2021), and the method has been extended to rock-salt and perovskite compounds, proposed as a cheap feature for machine-learning materials screening, and adapted in spirit to other bulk-reference schemes (the vacuum level at pore centers in metal-organic frameworks, and internal references for bulk band alignment). The same 2021 assessment compares the interstitial-potential charges with Bader and Phillips ionicity without providing a quantitative reconciliation, and its accuracy is reported to decrease for compounds with small lattice constants — so the two questions the 1977 paper left open are still open.

Problem Statement

The objective is a quantitative microscopic account of the electrostatic polarity of zinc-blende-type semiconductors, centered on the two issues that the 1977 Frensley-Kroemer analysis left open and that subsequent literature, including the 2021 first-principles revival, has documented but not resolved. (The interstitial method has since been applied to related binary and mixed families — rock-salt, perovskite — which provide useful context but are not part of the required objective.)

Component 1 — reconciliation of effective-charge measures. Provide a microscopic theory (first-principles and/or analytic) that quantitatively relates the interstitial-potential (static) effective charges to the other operational charge metrics for the same compounds: the dynamic (transverse/Szigeti and Born) charges from infrared spectroscopy and density-response calculations, LCAO/tight-binding population charges, Bader density-partition charges, and the Phillips spectroscopic ionicity. The theory should account for the documented pattern of disagreement among these metrics — including the comparisons reported by the 2021 first-principles assessment between Frensley-Kroemer-derived ionicity and Bader and Phillips ionicity — and should determine, for each pair of metrics, whether the discrepancy is a convention artifact (removable by a calculable transformation) or a genuine physical difference (each metric measuring a different observable). The theory should provide the calculable transformation for pairs whose discrepancy is a convention artifact and, for pairs reflecting a genuine physical difference, the physical reason for the difference together with its magnitude. The metric set over which these pairs are formed is the five operational charge metrics defined in the Background — the static interstitial charge, the dynamic charges (with the Szigeti and Born charges both addressed within that single family), LCAO population charges, Bader charges, and the Phillips ionicity.

Component 2 — error theory of the interstitial band reference. Derive, from the microscopic charge density, the error of the interstitial-potential absolute band-energy reference: the contributions of the neglected interstitial charge density, of the non-spherical bond (valence) charge that crowds the tetrahedral site, of the finite sampling volume around each interstitial site, of the pseudopotential/functional approximation, and of the ionic-model dipole correction when the reference is carried to a heterojunction or a surface. The theory should be predictive rather than merely descriptive, accounting for the documented empirical limitations of the reference — in particular the reduced accuracy for compounds with small lattice constants reported by the 2021 assessment — and for the sensitivity of the point-ion interface dipole to the choice of electronegativity scale and to pseudopotential quality, notably for the Al- and Cd-containing compounds (AlAs, CdS, CdSe) singled out by the original authors. The five listed contributions define the required scope of the error theory: each must be accounted for, and the theory should quantify the magnitude of each contribution rather than merely naming the sources of error, whether the five are covered in a single self-consistent treatment or in separate ones.

References

  1. W. R. Frensley and H. Kroemer, 'Prediction of semiconductor heterojunction discontinuities from bulk band structures,' J. Vac. Sci. Technol. 13, 810 (1976). DOI: 10.1116/1.568995, https://doi.org/10.1116/1.568995
  2. W. R. Frensley and H. Kroemer, 'Interstitial potential differences, electronegativity differences, and effective ionic charges in zinc-blende-type semiconductors,' Appl. Phys. Lett. 31, 48 (1977). DOI: 10.1063/1.89475, https://doi.org/10.1063/1.89475
  3. W. R. Frensley and H. Kroemer, 'Theory of the energy-band lineup at an abrupt semiconductor heterojunction,' Phys. Rev. B 16, 2642 (1977). DOI: 10.1103/PhysRevB.16.2642, https://doi.org/10.1103/PhysRevB.16.2642
  4. W. A. Harrison, 'Elementary theory of heterojunctions,' J. Vac. Sci. Technol. 14, 1016 (1977). DOI: 10.1116/1.569312, https://doi.org/10.1116/1.569312
  5. J. St. John and A. N. Bloch, 'Quantum-Defect Electronegativity Scale for Nontransition Elements,' Phys. Rev. Lett. 33, 1095 (1974). DOI: 10.1103/PhysRevLett.33.1095, https://doi.org/10.1103/PhysRevLett.33.1095
  6. W. R. Frensley, 'Heterostructure and Quantum Well Physics,' in Heterostructures and Quantum Devices (VLSI Electronics: Microstructure Science, Vol. 24, N. G. Einspruch and W. R. Frensley, eds., Academic Press, 1994), Ch. 1, pp. 1-24. DOI: 10.1016/B978-0-12-234124-3.50006-9, https://doi.org/10.1016/B978-0-12-234124-3.50006-9
  7. C. G. Van de Walle and R. M. Martin, 'Theoretical study of band offsets at semiconductor interfaces,' Phys. Rev. B 35, 8154 (1987). DOI: 10.1103/PhysRevB.35.8154, https://doi.org/10.1103/PhysRevB.35.8154
  8. J. Tersoff, 'Band Lineups at II-VI Heterojunctions: Failure of the Common-Anion Rule,' Phys. Rev. Lett. 56, 2755 (1986). DOI: 10.1103/PhysRevLett.56.2755, https://doi.org/10.1103/PhysRevLett.56.2755
  9. W. R. L. Lambrecht and B. Segall, 'Interface-bond-polarity model for semiconductor heterojunction band offsets,' Phys. Rev. B 41, 2832 (1990). DOI: 10.1103/PhysRevB.41.2832, https://doi.org/10.1103/PhysRevB.41.2832
  10. E. T. Yu, J. O. McCaldin, and T. C. McGill, 'Band Offsets in Semiconductor Heterojunctions,' in Solid State Physics, Vol. 46, H. Ehrenreich and D. Turnbull, eds. (Academic Press, 1992), pp. 1-146. DOI: 10.1016/S0081-1947(08)60397-5, https://doi.org/10.1016/S0081-1947(08)60397-5
  11. R. T. Tung, 'Recent advances in Schottky barrier concepts,' Mater. Sci. Eng. R 35, 1 (2001). DOI: 10.1016/S0927-796X(01)00037-7, https://doi.org/10.1016/S0927-796X(01)00037-7
  12. D. K. Ferry, 'Evolution of Physics and Chemistry of Surfaces and Interfaces: A Perspective of the Last 40 Years,' J. Vac. Sci. Technol. B 31, 048501 (2013). DOI: 10.1116/1.4806761, https://doi.org/10.1116/1.4806761
  13. T. Higuchi and H. Y. Hwang, 'General considerations of the electrostatic boundary conditions in oxide heterostructures,' in Multifunctional Oxide Heterostructures (E. Y. Tsymbal, E. R. A. Dagotto, C.-B. Eom, and R. Ramesh, eds., Oxford University Press, 2012). DOI: 10.1093/acprof:oso/9780199584123.003.0007, https://doi.org/10.1093/acprof:oso/9780199584123.003.0007
  14. K. T. Butler, C. H. Hendon, and A. Walsh, 'Electronic Chemical Potentials of Porous Metal-Organic Frameworks,' J. Am. Chem. Soc. 136, 2703 (2014). DOI: 10.1021/ja4110073, https://doi.org/10.1021/ja4110073
  15. Y. Kumagai, K. T. Butler, A. Walsh, and F. Oba, 'Theory of ionization potentials of nonmetallic solids,' Phys. Rev. B 95, 125309 (2017). DOI: 10.1103/PhysRevB.95.125309, https://doi.org/10.1103/PhysRevB.95.125309
  16. D.-H. Choe, D. West, and S. Zhang, 'Band Alignment and the Built-in Potential of Solids,' Phys. Rev. Lett. 121, 196802 (2018). DOI: 10.1103/PhysRevLett.121.196802, https://doi.org/10.1103/PhysRevLett.121.196802
  17. L. Harnett-Caulfield and A. Walsh, 'Assessment of interstitial potentials for rapid prediction of absolute band energies in crystals,' J. Chem. Phys. 155, 024113 (2021). DOI: 10.1063/5.0044866, https://doi.org/10.1063/5.0044866 (Open access.)
  18. N. Bhattacharya, A. Sen, K. Qu, A. Sinha, R. K. Patel, S. Kumar, J. Zhang, P. Mandal, S. C. Joshi, S. K. Ojha, J. Maity, Z. Zhang, H. Zhou, F. Rodolakis, P. Shafer, C. Klewe, J. W. Freeland, Z. Yang, U. V. Waghmare, and S. Middey, 'Site-Selective Polar Compensation of Mott Electrons in a Double-Perovskite Heterointerface,' Phys. Rev. Lett. 134, 176201 (2025). DOI: 10.1103/PhysRevLett.134.176201, https://doi.org/10.1103/PhysRevLett.134.176201