weinberg_1979#
Z. A. Weinberg, G. W. Rubloff and E. Bassous, “Transmission, photoconductivity, and the experimental band gap of thermally grown SiO2 films”, Phys. Rev. B 19, 3107 (1979)
Source record · Catalogue · Review register
Source kind: measurement
Recorded technique: Thermally grown amorphous SiO2 on (100) silicon (dry O2 at 1000 degrees C, dry O2 with 4.5 % HCl, or wet oxidation at 800 degrees C). Transmission 7-14 eV through free membranes (436, 1100 and 5110 A; Si3N4-protected anisotropic etching of the substrate; thickness +-1 % by ellipsometry), McPherson 225 monochromator with an H2 lamp, 8 A resolution, sodium-salicylate detector; the absorption coefficient from the exact thin-film transmission formula (Appendix), iterated from Philipp’s k with Philipp’s n fixed. Photoconductivity of Al-SiO2-Si capacitors (575-3500 A; Al 160 A thick, a masked and shielded contact), its spectrum and its dependence on the field at fixed photon energy (2.6-11.1 eV). Diodes under 450-1350 A oxide for photoluminescence and excitons (pp. 3109-3115).
Materials in structured entries: SiO2, SiO2_quartz
Dataset quantities: band_gap
Contents: 4 datasets; 0 formulas.
Curation notes#
Pages 3107-3117 are pdf pages 1-11. Identity checked against the first page.
BAND GAP (the paper’s result). ‘we associate the change in curvature seen in Fig. 9 with the SiO2 band gap and deduce that E_g ~ 9.3 eV. This association and the 9.3-eV value are, obviously, subject to a more detailed modeling of the problem’ (p. 3114): the transition energy in the field dependence of the photocurrent (yield against field, curvature change as for internal photoemission over a barrier), proposed as ‘the most direct experimental estimate of the band gap’ of thermally grown (amorphous) SiO2, direct or indirect not specified (p. 3115). Band-structure values the paper cites for alpha-quartz: indirect 9.2 eV and lowest direct 9.8 eV (Chelikowsky and Schluter); 9.8 eV direct for idealised beta-cristobalite (Ciraci and Batra) (p. 3107): theory, notes only.
ABSORPTION EDGE (figures only; not digitised). Fig. 3 transmission of the three membranes; Fig. 4 the absorption coefficient 7-14 eV of the 436-A film against Philipp’s Kramers-Kronig alpha (shifted by +0.075 eV to align the first peak, at 10.45 eV) and Powell and Morad’s 580-A film; Fig. 5 the tail with Appleton et al.’s points. ‘The accuracy of alpha in our data is estimated to be better than +-2% for the region where T < 0.2, which applies from about 9.2 eV for the 5110-A film to about 11.3 eV for the 436-A film. Above 11.3 eV the scattered light correction becomes appreciable and the error could be as high as 10%. The reproducibility of alpha among membranes ... has been within +-5%’ (p. 3111); reliable only above ~9.2 eV (p. 3111). The tail ‘extends down to at least 8.4 eV’ and is ‘approximately exponential ... in the range ~9.2 eV < h nu < ~10.2 eV, corresponding to about 2 decades in alpha’ (Urbach-like; p. 3115). Philipp’s Table X k at 0.11-0.13 um is taken from this work (philipp_1985/weinberg_k, the only tabulation). Figs. 6-9: photoconductivity spectra (575, 1015, 2310 A; field 2e6 V/cm) and the field dependence at 2.6-11.1 eV. Photoluminescence yield upper limit ~1e-4 at 10.2 eV (p. 3115). Barriers: Al Fermi level to SiO2 conduction band 3.15 eV; Si to SiO2 conduction band 3.1-3.2 eV (pp. 3107, 3114).
REVIEW OF EARLIER VALUES (pp. 3107-3108, 3115), as the paper states them. Those with a single value and a clear material and kind are the reproduced datasets below; the others are recorded here only: - Goodman: hole photoemission from Si into thermal SiO2, 4.9 eV to the SiO2 valence band, with the 3.1-3.2 eV conduction-band offset ‘a band gap of 8.0-8.1 eV’ (a range, and photoemission: the vocabulary’s band-gap kinds have no photoemission). - DiStefano and Eastman: besides the 9.0 eV photoconductivity gap (dataset), ‘8.9 +- 0.2 eV from photoemission of electrons from SiO2 into vacuum’ (threshold 10.2 eV); ‘revised upward to 9.3 eV if one assumes that the valence-band edge of SiO2 lies at 10.6 eV below vacuum’ (Ibach and Rowe). - Williams: 8 eV, estimated from an early absorption measurement ‘in quartz’ (Ref. 16) to 8.44 eV (crystalline or fused not said). - Powell and Derbenwick: besides the ~9 eV edge (dataset), ‘a band gap of 8.8 eV from their measurements of hole trapping in thermally grown SiO2 films illuminated by vuv photons’ (a charging threshold; no kind fits). - Stephenson and Binkowski (XPS): limits 5.55-7.8 eV (alpha-quartz), 5.05-7.3 eV (fused silica), up to 8.3 eV (OH-rich silica); criticised (defects from vacuum milling). - Zakis et al.: electron-hole creation ‘above 12 eV’. - Appleton et al. (alpha-quartz and vitreous SiO2), following Mott: ‘a direct band gap of 10.6 eV with the possibility of a lower indirect band gap of about 9 eV’ (a suggestion). - Mott: first ‘about 10.6 eV’, later ‘about 9 eV’, indirect, with 10.45 eV a direct-gap exciton (review). - Kaminow et al.: the absorption edge of their dry silica 0.2 eV above that of wet silica.
Recorded bibliographic identity status: VALIDATED_WITH_WARNINGS. This concerns identification of the source; it does not establish the scientific correctness of its contents.
Datasets#
weinberg_1979/band_gap#
{
"id": "band_gap",
"material": "SiO2",
"table": "text",
"quantity": "band_gap",
"unit": "eV",
"abscissa": "none",
"use": "primary",
"parameters": {
"kind": "photoconductivity",
"form": "film",
"technique": "field dependence of photoconductivity"
}
}
1 inline points in the source record.
E_g ~ 9.3 eV of thermally grown (amorphous) SiO2, from the field dependence of the photoconductivity of Al-SiO2-Si structures (Fig. 9; abstract p. 3107, Sec. II C p. 3114, Sec. III p. 3115). ‘subject to a more detailed modeling’; no uncertainty stated.
weinberg_1979/literature_photoconductivity#
{
"id": "literature_photoconductivity",
"material": "SiO2",
"table": "text",
"quantity": "band_gap",
"unit": "eV",
"abscissa": "label",
"use": "reproduced",
"reproduced_from": [
"T. H. DiStefano and D. E. Eastman, Solid State Commun. 9, 2259 (1971)",
"A. N. Trukhin, Phys. Status Solidi B 86, 67 (1978)"
],
"parameters": {
"kind": "photoconductivity"
}
}
3 inline points in the source record.
Earlier photoconductivity gaps as the paper states them (pp. 3107-3108). DiStefano and Eastman: ‘an interband gap of 9.0 eV by fitting their photoconductivity spectra (performed on a 5000-A thermally grown SiO2 film) to a parabolic absorption-edge characteristic of an indirect gap’ (the paper finds the fit unreliable because of interference, p. 3112). Trukhin: photoconductivity of crystalline quartz, fused silica and thermally grown films, ‘a direct band gap of 12 eV and a lowest band gap of 9.5 eV’; the paper judges these spectra erroneous (LiF-covered samples, inadequate shielding, p. 3113).
weinberg_1979/literature_optical#
{
"id": "literature_optical",
"material": "SiO2",
"table": "text",
"quantity": "band_gap",
"unit": "eV",
"abscissa": "label",
"use": "reproduced",
"reproduced_from": [
"R. J. Powell and M. Morad, J. Appl. Phys. 49, 2499 (1978)",
"R. J. Powell and G. J. Derbenwick, IEEE Trans. Nucl. Sci. 18, 99 (1971)"
],
"parameters": {
"kind": "optical"
},
"uncertainty": {
"column": true,
"level": "not stated"
}
}
2 inline points in the source record.
Earlier absorption-edge gaps as the paper states them (pp. 3108, 3115). Powell and Morad: ‘a value of 8.0 +- 0.2 eV, which is the same threshold they find in the absorption coefficient, photoconductivity spectra, and positive charging effects of thermally grown SiO2 films’ (a square-root fit of the absorption tail; the paper calls it ‘very doubtful’, p. 3112). Powell and Derbenwick ‘used Philipp’s results to conclude that the fundamental absorption edge of SiO2 is at approximately 9 eV’. The uncertainty level of +-0.2 is not stated.
weinberg_1979/literature_quartz#
{
"id": "literature_quartz",
"material": "SiO2_quartz",
"table": "text",
"quantity": "band_gap",
"unit": "eV",
"abscissa": "label",
"use": "reproduced",
"reproduced_from": [
"E. Loh, Solid State Commun. 2, 269 (1964)"
],
"parameters": {
"kind": "optical"
}
}
1 inline points in the source record.
‘The earliest reflectance spectrum of SiO2 (in the form of natural brazilian quartz) was reported by Loh; he identified the first spectral peak (~10.4 eV) as excitonic and crudely estimated the band gap to be above it, at about 11 eV’ (pp. 3107-3108). Axis not stated.
Formulas#
No formulas are recorded.
Corrections#
No dataset errata are recorded in this source record.
Formula corrections, where recorded, appear with their formulas above. Curation notes may also discuss corrections or unresolved interpretation. Consult the review register for the checks and decisions.