ditchburn_1953#
R. W. Ditchburn and D. W. O. Heddle, “Absorption cross-sections in the vacuum ultra-violet. I. Continuous absorption of oxygen (1800 to 1300 A)”, Proc. R. Soc. Lond. A 220, 61 (1953)
Source record · Catalogue · Review register
Source kind: measurement
Recorded technique: Photographic absorption spectrophotometry (Ilford Q1 plates) with a 1 m concave-grating vacuum spectrograph (a small fluorite Cario-Schmidt-Ott spectrograph in preliminary work) and a water-cooled capillary hydrogen discharge (many-line molecular spectrum); intensity reduced in a known ratio by grids (transmission calibrated photographically against a rotating sector); stray radiation reduced to ‘a negligible fraction’ by diaphragms. Paths 205 +- 1 cm (the spectrograph filled with O2, 1-12e-3 mm Hg) and 0.043 +- 0.0025 cm (a brass cell with fluorite windows, 5-200 mm Hg); O2 from cylinders (CO2 and water removed) or from heated permanganate. About fifty emission lines between 1360 and 1670 A, on average ten measurements each; Beer’s law tested with path ratio 5000:1 and pressure ratio 200000:1, ‘No systematic deviations’ (pp. 62-67).
Materials in structured entries: O2
Dataset quantities: absorption_cross_section, oscillator_strength
Contents: 5 datasets; 0 formulas.
Curation notes#
Journal pages 61-70 = PDF pages 2-11 (page 1 is the JSTOR cover; page 2 opens with the end of the preceding paper’s references). Identity checked against the first page (PDF page 2): Ditchburn and Heddle, Proc. R. Soc. A 220, 61, received 24 April 1953. O2 only.
Quantity (Eq. 1, p. 61): ln(I0/I) = alpha p l = N sigma_nu p l, ‘p is the pressure (reduced to 0 C) in atmospheres and N = 2.705 x 10^19 cm^-3’: alpha (cm^-1) is the natural-log absorption coefficient at 1 atm and 0 C. The datasets keep the printed alpha with scale = 1e18 / 2.705e19 = 0.0369686 to the cross section in Mb, the paper’s own conversion: its maximum alpha 490 cm^-1 ‘corresponds’ to 1.81e-17 cm^2 (490 / 2.705e19 = 1.811e-17). The ideal-gas Loschmidt density at 0 C and 1 atm, 2.6868e19 cm^-3, is 0.67 % lower than the paper’s N, so the physical cross section alpha / n0 would be 0.68 % larger than these values (well inside the stated errors; noted for review). Temperature of the gas not stated (room temperature). Wave numbers are vacuum wave numbers (69000 cm^-1 = 1449 A, the paper’s ‘1450 A’).
Accuracy (p. 69): ‘the average probable error of a single observation is rather less than 60 cm^-1. The limits of error given in the table are based on the assumption that the average of n observations is subject to an error of +-60 n^-1/2 cm^-1. The observational error calculated in this way is probably a little too high and is certainly correct within a factor of 2.’ Check (2026-09-25): the printed errors of Table 1 do not follow 60 n^-1/2; they follow about 80 n^-1/2 (n = 12: 23; 24: 16; 74: 9; 56: 11; 64: 10; 21: 17; 39: 13; 23: 17; 20, 19: 18), with 20 printed for every group of 15-17 observations except the 64108 group (16 observations, printed +-16 where 80 n^-1/2 gives 20). The printed errors are kept as the paper’s probable errors.
Check of the group means (2026-09-25): the observation-weighted means of each group’s lines (columns 2 and 3 with the bracketed numbers of observations) reproduce columns 4 and 5 for 20 of the 21 groups (to 0.5 cm^-1 in alpha; the 69484 group’s wave number comes out 69500). The group 64060-63810 (306 (7), 307 (6), 228 (3)) gives 63919 cm^-1 and 291.8 cm^-1 against the printed 64108 and 277 +- 16: a printed mean wave number above every line of its group is impossible for a weighted mean, so the row as printed is inconsistent; no single misprint explains both columns and no erratum is made. Doubt queued (TC-213).
Text values (pp. 61, 67, 70): the maximum ‘is 490 cm^-1, and the corresponding maximum cross-section is 1.81 +- 0.1 x 10^-17 cm^2 at 69000 cm^-1 and 1450 A’; half-value at 63800 and 73000 cm^-1 (abstract: 1567 and 1370 A, not data). Note added in proof (p. 70): ‘Making a reasonable extrapolation of the curve in figure 4, we obtain an f-value of 0.215 +- 0.020’ (against Watanabe et al.’s 0.161, watanabe_1953, and Ladenburg and Van Voorhis’s 0.193, ladenburg_1933). The smooth curve of Fig. 4 and the points of Schneider (1937) and Ladenburg and Van Voorhis (1933) plotted with it are figures only.
Comparison (p. 69): Watanabe, Inn and Zelikoff’s preliminary maximum 380 cm^-1 ‘at 71000 cm^-1’ (watanabe_1952, which prints 1420 A, 70420 cm^-1) is ‘25 % lower’ and ‘may be due to the presence of scattered radiation’; agreement with Ladenburg and Van Voorhis (ladenburg_1933) in the maximum and its position; below them, and with Schneider, at 60000-65000 cm^-1. Later photoelectric work (metzger_1964: 402 +- 20 cm^-1 at 1422 A) supports the lower maximum; these photographic values are about 20 % higher near the peak. Figure 3 (Beer’s law plots at 1435 and 1670 A) is a figure only. The band oscillator strengths of the Schumann-Runge system are part II (ditchburn_1954).
Recorded bibliographic identity status: VALIDATED_WITH_WARNINGS. This concerns identification of the source; it does not establish the scientific correctness of its contents.
Datasets#
ditchburn_1953/means#
{
"id": "means",
"material": "O2",
"file": "O2/ditchburn_1953__means.csv",
"table": "tbl_9b71613fa0c5",
"quantity": "absorption_cross_section",
"unit": "Mb",
"scale": 0.03696857671,
"abscissa_unit": "cm-1",
"wavelength_medium": "vacuum",
"use": "primary",
"uncertainty": {
"column": true,
"level": "probable error"
}
}
Table 1 columns 4-6: the mean wave number and mean alpha (cm^-1 at 1 atm, 0 C) of 21 groups of neighbouring lines (both paths), the points of Fig. 4; scale = 1e18 / 2.705e19 (the paper’s N) to Mb. sigma is the printed +- (probable error, ‘certainly correct within a factor of 2’; see the record’s notes: the printed errors are about 80 n^-1/2, not the stated 60 n^-1/2). The group ending at 63810 cm^-1 (printed 64108, 277 +- 16) does not reproduce from its lines (63919, 291.8): kept as printed, doubt queued. Temperature not stated (room temperature).
ditchburn_1953/long_path#
{
"id": "long_path",
"material": "O2",
"file": "O2/ditchburn_1953__long_path.csv",
"table": "tbl_9b71613fa0c5",
"quantity": "absorption_cross_section",
"unit": "Mb",
"scale": 0.03696857671,
"abscissa_unit": "cm-1",
"wavelength_medium": "vacuum",
"use": "check"
}
Table 1 column 2: alpha (cm^-1 at 1 atm, 0 C) on single hydrogen lines with the 205 cm path (1-12e-3 mm Hg), 15 lines; the number of observations per line (2-21) is kept in the printed column. scale = 1e18 / 2.705e19 to Mb. The same observations make up the means dataset: a check, not fitted beside it. Per-line error by the paper’s rule about 60 n^-1/2 cm^-1 (probable error; not printed per line).
ditchburn_1953/short_path#
{
"id": "short_path",
"material": "O2",
"file": "O2/ditchburn_1953__short_path.csv",
"table": "tbl_9b71613fa0c5",
"quantity": "absorption_cross_section",
"unit": "Mb",
"scale": 0.03696857671,
"abscissa_unit": "cm-1",
"wavelength_medium": "vacuum",
"use": "check"
}
Table 1 column 3: alpha (cm^-1 at 1 atm, 0 C) on single hydrogen lines with the 0.043 cm path (5-200 mm Hg), 44 lines; the number of observations per line (3-64) is kept in the printed column. scale = 1e18 / 2.705e19 to Mb. The same observations make up the means dataset: a check, not fitted beside it. Where both paths measured a line the values differ by up to 35 % (63810 cm^-1: 307 and 228), within the stated single-observation error of about 60 cm^-1.
ditchburn_1953/maximum#
{
"id": "maximum",
"material": "O2",
"table": "text",
"quantity": "absorption_cross_section",
"unit": "cm2",
"scale": 1e-17,
"abscissa_unit": "cm-1",
"wavelength_medium": "vacuum",
"use": "check",
"uncertainty": {
"absolute": 0.1,
"level": "not stated"
}
}
1 inline points in the source record.
p. 67 (and the abstract, p. 61): ‘The maximum value of alpha is 490 cm^-1, and the corresponding maximum cross-section is 1.81 +- 0.1 x 10^-17 cm^2 at 69000 cm^-1 and 1450 A’, read from the smooth curve of Fig. 4 through the means: a check on the means dataset, not independent. +-0.1e-17 as printed (level not stated).
ditchburn_1953/continuum_f#
{
"id": "continuum_f",
"material": "O2",
"table": "text",
"quantity": "oscillator_strength",
"abscissa": "label",
"use": "check",
"uncertainty": {
"absolute": 0.02,
"level": "not stated"
}
}
1 inline points in the source record.
p. 70, note added in proof: the oscillator strength of the Schumann-Runge continuum, ‘Making a reasonable extrapolation of the curve in figure 4, we obtain an f-value of 0.215 +- 0.020’; integration range and extrapolation not stated. Index abscissa: an integrated strength, not a line. Compare watanabe_1953 (0.161), metzger_1964 (0.142), ladenburg_1933 (0.193). Part II (ditchburn_1954) adds its band sum 0.044 to this for a total of 0.259. A check.
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.