windholz_1980#

L. Windholz, “Stark Effect of Ar I Lines”, Phys. Scr. 21, 67 (1980)

Source record · Catalogue · Review register

Source kind: measurement

Recorded technique: Static (dc) Stark effect of 91 Ar I lines (4100-6800 A, air) in fields up to about 700 kV/cm (nine photographic records with maximum fields 57.6-693.3 kV/cm): argon at about 5 x 10^-2 torr in the field between shaped electrodes, excited by a helium canal ray; a three-prism spectrograph (6.1 A/mm at 4100 A, 51.5 A/mm at 6700 A), the field along the slit marked by a wedge grid and calibrated by the shifts of helium lines; field-free reference spectra on each plate; the shifts measured with an Abbe comparator (Secs. 2-3). The shifts are attributed entirely to the upper levels (p. 68) and fitted, per level component, by Delta nu = a1 F^2/(a2 - Delta nu) (Eq. 3; Table III) or a polynomial in F (Eq. 5; Table IV).

Materials in structured entries: None recorded.

Dataset quantities: None recorded.

Contents: 0 datasets; 0 formulas.

Curation notes#

Context only, no datasets. Phys. Scr. 21, 67-74 (1980); the library PDF has an IOPscience cover page, so the article is PDF pages 2-9 (journal p. 67 = PDF p. 2; ‘p.’ below is the journal page). Identity checked against the first article page: journal, volume, pages, author, title (‘Stark Effect of Ar I’-Lines’ in the heading, ‘Stark effect of Ar I lines’ in the abstract and running head), received 24 April 1979, final version 20 June 1979. The library’s identity is VALIDATED_WITH_WARNINGS. The abstract’s own citation line prints ‘21, 61-14’ for 67-74 on the text layer; the page image reads 67-74.

WHICH LEVELS. The paper’s levels are the UPPER levels of 4s-5p lines (5p, 5p’) and of 4p-nd and 4p-ns lines (4d, 4d’, 5d, 5d’, 6d, 6d’, 7d, 8d, 6s, 7s, 7s’, 8s, 9s): 49 levels (Table II). The 4s and 4p levels are only lower levels: ‘it can be assumed that the lower levels do not change their energy in the field within the accuracy of measurement. Thus, shifts and splittings of the spectral lines must be interpreted as shifts and splittings of the upper levels’ (p. 68). The paper therefore gives nothing for the Ar 4s or 4p levels’ polarisabilities (OEP 8’s inputs).

LEVEL LABELS. Windholz writes nl (K) J with a prime for the 2P1/2 core, e.g. ‘5p (1/2) 0’, ‘4d’ (5/2) 3’, the Racah pair-coupling label nl[K]_J without brackets. Every level energy of Tables III and IV (cm^-1, one decimal; from Moore 1949, ref. 6) agrees with NIST’s (data/external/nist_asd/Ar_I_levels.tsv) to 0.1 cm^-1 for all 49 levels, and the label then maps to NIST’s 3s2.3p5.(2P*<3/2>) or (2P*<1/2>).nl 2[K] J (unprimed and primed); for example 5p (1/2) 0 at 117 563.0 = (2P*<3/2>).5p 2[1/2] J = 0 (Paschen 3p5); 5p’ (1/2) 0 at 118 870.9 = (2P*<1/2>).5p 2[1/2] 0 (3p1); 5d (1/2) 0 at 121 794.1 = (2P*<3/2>).5d 2[1/2]* 0. Table III’s two 4d’ (3/2) rows are printed with a single prime (the text layer reads a double prime on one; the image has one).

WHAT THE PAPER PRINTS. Table I (pp. 69-70): 91 lines (air wavelengths, transitions, number of components, F_max, relative intensity ratios). Table II (p. 69): the upper levels and their lines. Table III (pp. 70-71): per level and component (numbered by increasing shift; no assignment to m, ‘A quantum mechanical correlation between the magnetic quantum number m and the components could not be found’, p. 71), the level energy, F_max (kV/cm), a1 (printed unit ‘kV^-2’), a2 (cm^-1) and the Lindholm Stark constant C4 = c e^2 Delta nu/F^2 (Eq. 6, ‘CGS’) at F = 0 (the limit a1/a2) and at F = F_max. Delta nu is the level’s wavenumber change (cm^-1), F in kV/cm. Seven components marked ‘*’ (7d) were measured at one field only (Table V, F = 104.4 kV/cm: -26.5, -34.2, -36.9, -9.6 and -43.0 cm^-1). Table IV (p. 72): the 5d’ (3/2) 2, 6d (1/2) 0 and 6d (1/2) 1 levels, whose shifts turn over, as polynomial coefficients A1-A4 of Eq. 5 with their ranges of validity (no C4). Table VI (p. 73): line shifts of 47 lines at one field, compared with Minnhagen (Ark. Fys. 1, 425 (1950)), reproduced. Accuracy: the fitted shifts deviate from the measured ones by less than +-1 cm^-1 (+-2 cm^-1 for the 7d, 8d, 7s’, 8s and 9s groups, p. 69). Figs. 3-7: level diagrams of the 4d’, 5d, 5d’, 6d and 7d groups against F (shapes; the coefficients are printed).

NOT TAKEN (no quantity fits; Data rule 7). level_polarizability (TC-1011) is a level’s scalar, vector or tensor polarisability in au, SI, angstrom3, cm3 or kHz/(kV/cm)2. The paper prints none of these: (a) its constants are C4 in CGS (Hz cm^4, ordinary frequency) and fit coefficients a1, a2 of a non-quadratic law, so a polarisability is a derived number (alpha/h = -2 c a1/a2), not a printed one; (b) each value belongs to an unassigned Stark component, which is alpha_0 plus an unknown tensor part, neither the scalar nor the tensor component of the quantity, except for the J = 0 levels (5p (1/2) 0, 5p’ (1/2) 0, 5d (1/2) 0, 7d (1/2) 0), whose single component is scalar; (c) each is the line’s shift with the lower level’s shift set to zero by assumption, i.e. alpha(upper) - alpha(lower) attributed to the upper level; (d) none is a 4s or 4p level. For the record (derived here, not printed; the J = 0 scalars from C4 at F = 0): 5p (1/2) 0, alpha_0/h = 3590 kHz/(kV/cm)^2; 5p’ (1/2) 0, 2768; 5d (1/2) 0, 6719 (about 14 400, 11 100 and 27 000 au on werbowy_2026’s factor 0.2488319); across all 5p and 5p’ components 2079-9870, across the 4d to 9s components 2507 to 6.0 x 10^5. Table IV’s polynomials, Table V’s single shifts and Table VI’s line shifts are line or level shifts at a field, not a quantity (declined, TC-1011).

CHECKED (scratch script, 2026-10-12). Every C4 at F = 0 of Table III (65 components with a1, a2) equals c e^2 (a1/a2)/F_unit^2 with c and e in CGS and 1 kV/cm = 3.33564 statV/cm, within the last printed digit; every C4 at F = F_max equals c e^2 Delta nu(F_max)/F_max^2 with Delta nu from Eq. 4 (the root with Delta nu -> 0 at F = 0). So C4 is in cm^4 s^-1 with ordinary frequency, and the a1 unit is cm^-2 per (kV/cm)^2 (printed ‘kV^-2’). Every level energy against NIST (above). Tables III, IV and II were read on the page images: Table III’s ‘2.10e--01’ (7s’ (1/2) 1, a1) is 2.10e-01 on the image; Table II prints 6212.93 among the 5d’ (3/2) 2 lines where Table I (6215.93, 4p’ (3/2) 2 - 5d’ (3/2) 2) and the Fig. 2 caption print 6215.93: a misprint in Table II, outside any value taken.

FINDINGS AGAINST werbowy_2026 (TC-1007, TC-1018). No level in common: Werbowy’s are the 4s and 4p levels (the lower levels here). Werbowy’s fitted 4s alpha_0 (56-96 kHz/(kV/cm)^2) and 4p alpha_0 (95-327) imply lower-level shifts of 0.46-0.78 cm^-1 (4s) and 0.78-2.7 cm^-1 (4p) at 700 kV/cm, against Windholz’s +-1 cm^-1 fit accuracy: the assumption of unshifted lower levels holds for the 4s-5p lines and is marginal for 4p-nd lines observed at the highest fields; the upper levels’ coefficients are about 20-180 times the 4s values (5p) and 8 to over 1000 times the 4p values (4d-9s). Werbowy cites this paper (its ref. 9) but does not use its numbers.

FOR OEP 8. Nothing direct: the argon 4p levels’ polarisabilities are not measured here, and these are static values. Further Stark data cited and not in the library: Minnhagen, Ark. Fys. 1, 425 (1950) (Ar I Stark effect to 159 kV/cm); Ryde, Atoms and Molecules in Electric Fields (1976).

Recorded bibliographic identity status: VALIDATED_WITH_WARNINGS. This concerns identification of the source; it does not establish the scientific correctness of its contents.

Datasets#

No datasets are recorded.

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.