wiese_1989#
W. L. Wiese, J. W. Brault, K. Danzmann, V. Helbig and M. Kock, “Unified set of atomic transition probabilities for neutral argon”, Phys. Rev. A 39, 2461 (1989)
Source record · Catalogue · Review register
Source kind: compilation
Recorded technique: Critical evaluation of the Ar I 4s-4p and 4s-5p transition probabilities: 4s-4p scaled to Quadfasel and Helbig’s laser-induced-fluorescence lifetimes of the ten 4p levels, the individual lines the average of Nick and Helbig’s branching ratios and Shumaker and Popenoe’s emission data (Chang and Setser’s branching ratios where Nick and Helbig have none); 4s-5p from three wall-stabilised-arc emission sets normalised to A(430.010 nm) = 3.77e5 s-1. The scale of the 4s-5p array set by the authors’ own measurement: branching ratios of all 16 lines from 5p[5/2]2 (Fourier-transform spectrometer, hollow-cathode lamp, 400-6100 nm) with its LIF lifetime.
Materials in structured entries: Ar
Dataset quantities: radiative_lifetime, transition_probability
Contents: 4 datasets; 0 formulas.
Curation notes#
Identity: Physical Review A, volume 39, number 5, 1 March 1989, pp. 2461-2471, W. L. Wiese (NIST), J. W. Brault, K. Danzmann, V. Helbig, M. Kock, “Unified set of atomic transition probabilities for neutral argon” (first page of the scanned PDF; no text layer). NIST ASD’s Ar I transition probabilities cite this paper (its bibliographic key T5223, resolved 2026-10-09).
Taken (TC-1011’s quantities): the transition probabilities of Tables IV, VI and I (dataset notes) and the adopted lifetime of 5p[5/2]2. A file’s row is the printed air wavelength (nm; unique in each table; standard air: 811.531 nm against NIST’s vacuum 811.754 nm); x is the transition in the paper’s Racah notation, lower-upper (4s[3/2]2-4p[5/2]3: nl[K]J, a prime for the 2P1/2 core, the parity mark dropped), as printed; the misprinted labels are corrected by errata. The paper prints no oscillator strength, no gf and no statistical weights (an f converted from A would be derived).
NOT TAKEN: Table II (p. 2465, tbl_c3d2bc60d69f), the ten 4p lifetimes of other papers (Quadfasel and Helbig, Chang and Setser, Shumaker and Popenoe renormalised by 0.85, Musiol et al., Lilly’s theory): reproduced, and Table IV carries their scale. The authors’ own LIF lifetime of 5p[5/2]2, 152 +- 7 ns (p. 2463), enters the adopted 153 ns, which is taken. Table I’s branching ratios 100R: its A carry them.
Not taken: Table III (4s-4p A of five lines by Nick and Helbig, Nubbemeyer, Shumaker and Popenoe, 1e5 s-1), Table V (4s-5p emission sets of Wende, Bues et al., Jones and Wiese), Table VII (ratios of A between 4s-4p and 4s-5p lines), Fig. 2 (A(430.010) of 15 emission experiments): other papers’ values, and Table I’s level energies (reproduced; NIST holds the levels).
Levels: the paper’s Racah (jK) labels nl[K]_J, primed for the 2P1/2 core, are NIST’s configuration (2P*<3/2>).nl or (2P*<1/2>).nl, term 2[K] (with * for the odd 4s, 3d, 5s levels) and J. Paschen: 1s5, 1s4, 1s3, 1s2 = 4s[3/2]2, 4s[3/2]1, 4s’[1/2]0, 4s’[1/2]1; 2p1-2p10 = 4p’[1/2]0, 4p’[1/2]1, 4p’[3/2]2, 4p’[3/2]1, 4p[1/2]0, 4p[3/2]2, 4p[3/2]1, 4p[5/2]2, 4p[5/2]3, 4p[1/2]1 (descending energy; Table II lists the levels in this order and the text gives 2p4’s lifetime, 29.3 ns, as Table II’s 4p’[3/2]1; “5p[5/2]2 (3p8 in Paschen notation)”, p. 2462). Every line of Tables I, IV and VI was identified by its printed wavelength among NIST’s level pairs (air wavelengths from NIST’s levels; residual <= 0.0006 nm in Tables IV and VI but 750.384 nm, 0.0028).
Misprints (labels and level energies; no A affected; the labels of taken rows are corrected by the errata below): Table IV prints 4p[1/2]1 for 696.543 nm (4p’[1/2]1, as Table III) and 4p’[3/2]2 for 763.511 nm (4p[3/2]2); Tables II, IV and VI print every [5/2]2 level primed (4p’[5/2]2 at 978.450, 842.465 and 801.479 nm; 5p’[5/2]2 at 462.844, 430.010 and 419.071 nm), which cannot exist (K = 1/2, 3/2 on the 2P1/2 core) and the text writes 5p[5/2]2; Table VI prints 5p[1/2]1 for 394.898 nm (5p’[1/2]1; the 4s[3/2]2 group would list 5p[1/2]1 twice). Table I prints 5s[5/2]2 for 5s[3/2]2 (2831.415 nm), 5s[1/2]1 for 5s’[1/2]1 (4938.586 nm), E_i “99 143.80” for 93 143.80 (4s[3/2]2: only 93 143.80 gives the printed 419.071 nm) and “114 716.61” for 113 716.61 (3d[5/2]3: only 113 716.61 gives 3045.369 nm). Table IV’s 750.384 nm is 750.387 in Table III and from NIST’s levels. Table V’s footnote d gives the lifetime-branching-ratio A of 462.844 nm as 0.0383 “(see Table II)”: Table I has 0.0396, and 0.0383 is Table VI’s recommended value, which footnote a says includes it.
Comparison with NIST ASD (TC-983, Ar_I_lines_nir.tsv, vacuum wavelengths): NIST holds 29 of Table IV’s lines with an A (all but the <0.001 one), 28 from this paper (T5223) and 1148.811 nm from Shumaker and Popenoe renormalised (T1242n, 1.9e5 s-1, the same as Table IV’s 0.19e6). Ratio Table IV / NIST: mean 0.9996, mean |deviation| 0.17 %, largest 1.1 % (750.384 nm, 44.5 against 4.5e7): every difference is NIST’s rounding to two or three digits. NIST’s grades follow the paper’s uncertainties (+-5 % B+, +-8 % and +-10 % B, +-15 % C+). Table VI against NIST: 28 lines, mean 0.9987, largest 1.1 %, all within NIST’s rounding. The agreement is NIST quoting this paper, not an independent check. The radiative widths sum A per 4p level give 1/sum A = 22.35 (2p1), 28.39, 28.93, 30.17, 24.88, 29.04, 29.70, 31.01, 30.21, 39.22 ns (2p10), within 3 % of Quadfasel and Helbig’s lifetimes (Table II) the scale was set by (the 4p levels decay only to 4s, p. 2464). The cells were checked against the table images one by one (2026-10-09); tools/extract/wiese_1989.py checks the labels against NIST’s levels.
Recorded bibliographic identity status: VALIDATED_WITH_WARNINGS. This concerns identification of the source; it does not establish the scientific correctness of its contents.
Datasets#
wiese_1989/A_4s_4p#
{
"id": "A_4s_4p",
"material": "Ar",
"file": "Ar/wiese_1989__A_4s_4p.csv",
"table": "tbl_c910a03b3fcc",
"quantity": "transition_probability",
"unit": "1/s",
"scale": 1000000.0,
"abscissa": "label",
"use": "primary",
"excluded_rows": {
"857.806": "printed <0.001: an upper bound (4s'[1/2]1-4p[1/2]0, J = 1 to J = 0 across the cores), not a value"
},
"uncertainty": {
"relative_rows": {
"750.384": 0.08,
"826.452": 0.05,
"840.821": 0.05,
"852.144": 0.08,
"922.450": 0.1,
"935.422": 0.08,
"978.450": 0.08,
"1148.811": 0.3,
"772.421": 0.05,
"794.818": 0.08,
"866.794": 0.08,
"1047.005": 0.15,
"667.728": 0.1,
"727.293": 0.05,
"738.398": 0.08,
"747.117": 0.08,
"751.465": 0.08,
"800.616": 0.15,
"810.369": 0.05,
"842.465": 0.05,
"965.778": 0.08,
"696.543": 0.05,
"706.722": 0.08,
"714.704": 0.08,
"763.511": 0.08,
"772.376": 0.05,
"801.479": 0.08,
"811.531": 0.08,
"912.297": 0.08
},
"level": "not stated"
},
"parameters": {
"notation": "racah"
}
}
Table IV (p. 2467), ‘Recommended atomic transition probabilities for the Ar I 4s-4p transition array’, A (1e6 s-1), all 30 lines (the 4p levels decay only to 4s, p. 2464): the averages of Nick and Helbig’s branching ratios and Shumaker and Popenoe’s emission data, both scaled to Quadfasel and Helbig’s lifetimes (Chang and Setser’s branching ratios for 2p8 and 2p10 lines Nick and Helbig did not measure). Per-row ‘Estimated uncertainty’ (+-5 to +-30 %): ‘our accuracy estimates, which include the uncertainties given for the lifetime and averaged emission data’ (p. 2466), level not stated. Rows are the printed air wavelengths; 750.384 nm is printed 750.387 in Table III and is 750.387 from NIST’s levels (a misprint of the row’s wavelength, which no erratum can correct; the line is not in doubt). Paschen: 1s2, 1s3, 1s4, 1s5 = 4s’[1/2]1, 4s’[1/2]0, 4s[3/2]1, 4s[3/2]2; 2p1-2p10 as in the record’s notes.
wiese_1989/A_4s_5p#
{
"id": "A_4s_5p",
"material": "Ar",
"file": "Ar/wiese_1989__A_4s_5p.csv",
"table": "tbl_9316afb55907",
"quantity": "transition_probability",
"unit": "1/s",
"scale": 1000000.0,
"abscissa": "label",
"use": "primary",
"uncertainty": {
"relative_rows": {
"425.936": 0.06,
"433.534": 0.12,
"433.356": 0.07,
"434.517": 0.1,
"451.073": 0.07,
"458.929": 0.2,
"459.610": 0.09,
"462.844": 0.06,
"470.232": 0.09,
"418.188": 0.09,
"419.103": 0.15,
"442.400": 0.15,
"452.232": 0.08,
"404.596": 0.1,
"404.442": 0.09,
"405.453": 0.15,
"419.832": 0.08,
"426.629": 0.12,
"427.217": 0.08,
"430.010": 0.05,
"436.379": 0.1,
"394.898": 0.09,
"394.750": 0.15,
"415.859": 0.09,
"416.418": 0.09,
"419.071": 0.08,
"420.067": 0.07,
"425.118": 0.15
},
"level": "2 sigma"
},
"parameters": {
"notation": "racah"
}
}
Table VI (p. 2469), ‘Recommended atomic transition probabilities for the Ar I 4s-5p transition array’, A (1e6 s-1): the 28 lines with a value (397.972 nm has none; 395.7 nm, 4s[3/2]2-5p’[3/2]1, ‘Line not observed’). The scale is the mean of the authors’ lifetime-branching-ratio A(430.010 nm) and the emission analysis’ (3.77e5 s-1), the lines the average of three emission sets normalised to it (p. 2470); footnote a (462.844 nm) ‘Includes the result of our lifetime-branching-ratio approach’, footnote c (419.071 nm) applies that result (Table I) instead of Bues et al.’s. Uncertainties ‘given, with two standard deviations’ (p. 2470). Rows are the printed air wavelengths.
wiese_1989/A_5p52_branches#
{
"id": "A_5p52_branches",
"material": "Ar",
"file": "Ar/wiese_1989__A_5p52_branches.csv",
"table": "tbl_62e5805103d7",
"quantity": "transition_probability",
"unit": "1/s",
"scale": 1000000.0,
"abscissa": "label",
"use": "check",
"uncertainty": {
"relative_rows": {
"419.071": 0.08,
"430.010": 0.07,
"462.844": 0.08,
"1929.491": 0.07,
"2056.879": 0.04,
"2512.509": 0.03,
"2797.740": 0.04,
"2831.415": 0.05,
"2978.811": 0.08,
"3045.369": 0.1,
"3505.799": 0.04,
"4239.098": 0.05,
"4556.209": 0.05,
"4591.384": 0.06,
"4938.586": 0.07
},
"level": "2 sigma"
},
"parameters": {
"notation": "racah"
}
}
Table I (p. 2464): the authors’ measured branching ratios of the 16 lines from 5p[5/2]2 (3p8, 116 999.39 cm-1; Fourier-transform spectrometer, hollow-cathode lamp) ‘and corresponding transition probabilities, normalized to the average (153 ns)’ of the lifetimes of Ralfs and Stryla et al.: the 15 lines with an A (6123.946 nm, 3d’[3/2]1, has 100R < 0.06 and none). use = check: Table VI takes over the three 4s-5p lines (419.071 nm the same 0.280; 430.010 nm 0.380 against 0.377; 462.844 nm 0.0396 against 0.0383), and the 12 5p-3d and 5p-5s lines (1929-4939 nm) are one measurement on one level; they are the only values of these lines in the library. The relative uncertainty of each row is the printed Delta R (two standard deviations, p. 2463); the lifetime’s +-7 ns (4.6 %, two standard deviations with a systematic allowance) is common to all rows and not included. Every A is 100R / 153 ns within the rounding but 3045.369 nm (0.148 printed; 2.25 % gives 0.1471). x is lower-upper. The printed lower-level energies E_i are not taken; two are misprinted (record notes).
wiese_1989/lifetime_5p52#
{
"id": "lifetime_5p52",
"material": "Ar",
"table": "text",
"quantity": "radiative_lifetime",
"unit": "ns",
"abscissa": "label",
"use": "primary",
"uncertainty": {
"column": true,
"level": "2 sigma"
},
"parameters": {
"method": "measured: laser-induced fluorescence from the metastable 4s[3/2]2, delayed coincidence, extrapolated to zero pressure; the mean of two measurements with one apparatus"
}
}
1 inline points in the source record.
Text p. 2464: the radiative lifetime of 5p[5/2]2 (3p8; NIST (2P*<3/2>).5p 2[5/2] J = 2): ‘The arithmetic mean of the two closely agreeing results, tau=153 nm [sic: ns], was therefore adopted for this work. An overall uncertainty of +-7 ns was obtained by combining the statistical measurement errors (two standard deviations) with an allowance for systematic errors.’ The two results: the authors’ LIF measurement (Ralfs, Ref. 56), 152 +- 7 ns, extrapolated to zero pressure over 0.026-0.26 Pa (p. 2463), and Stryla et al.’s 154 +- 7 ns with the same apparatus at lower densities (Ref. 57). The value normalises Table I.
Formulas#
No formulas are recorded.
Corrections#
[
{
"dataset": "A_4s_4p",
"row": "978.450",
"column": "x",
"printed": "4s'[1/2]1-4p'[5/2]2",
"corrected": "4s'[1/2]1-4p[5/2]2",
"reason": "A [5/2] level on the 2P1/2 core cannot exist (a p electron on j_c = 1/2 has K = 1/2 or 3/2): the level is 4p[5/2]2 (2p8), printed primed throughout Tables II and IV.",
"evidence": "The printed wavelength is NIST's air wavelength of (2P*<3/2>).4p 2[5/2] J=2 to this lower level (within 0.0005 nm; no other Ar I level pair within 0.8 nm); the text writes the analogous 5p level unprimed, '5p[5/2]2 (3p8 in Paschen notation)' (p. 2462)."
},
{
"dataset": "A_4s_4p",
"row": "842.465",
"column": "x",
"printed": "4s[3/2]1-4p'[5/2]2",
"corrected": "4s[3/2]1-4p[5/2]2",
"reason": "A [5/2] level on the 2P1/2 core cannot exist (a p electron on j_c = 1/2 has K = 1/2 or 3/2): the level is 4p[5/2]2 (2p8), printed primed throughout Tables II and IV.",
"evidence": "The printed wavelength is NIST's air wavelength of (2P*<3/2>).4p 2[5/2] J=2 to this lower level (within 0.0005 nm; no other Ar I level pair within 0.8 nm); the text writes the analogous 5p level unprimed, '5p[5/2]2 (3p8 in Paschen notation)' (p. 2462)."
},
{
"dataset": "A_4s_4p",
"row": "801.479",
"column": "x",
"printed": "4s[3/2]2-4p'[5/2]2",
"corrected": "4s[3/2]2-4p[5/2]2",
"reason": "A [5/2] level on the 2P1/2 core cannot exist (a p electron on j_c = 1/2 has K = 1/2 or 3/2): the level is 4p[5/2]2 (2p8), printed primed throughout Tables II and IV.",
"evidence": "The printed wavelength is NIST's air wavelength of (2P*<3/2>).4p 2[5/2] J=2 to this lower level (within 0.0005 nm; no other Ar I level pair within 0.8 nm); the text writes the analogous 5p level unprimed, '5p[5/2]2 (3p8 in Paschen notation)' (p. 2462)."
},
{
"dataset": "A_4s_4p",
"row": "696.543",
"column": "x",
"printed": "4s[3/2]2-4p[1/2]1",
"corrected": "4s[3/2]2-4p'[1/2]1",
"reason": "The upper level is 4p'[1/2]1 (2p2), not 4p[1/2]1 (2p10, which is the 912.297 nm line of the same group).",
"evidence": "Table III prints 4s[3/2]2-4p'[1/2]1 for 696.543 nm; NIST's air wavelength of (2P*<1/2>).4p 2[1/2] J=1 - 4s[3/2]2 is 696.543 nm, and 4p[1/2]1 - 4s[3/2]2 is 912.297 nm."
},
{
"dataset": "A_4s_4p",
"row": "763.511",
"column": "x",
"printed": "4s[3/2]2-4p'[3/2]2",
"corrected": "4s[3/2]2-4p[3/2]2",
"reason": "The upper level is 4p[3/2]2 (2p6), not 4p'[3/2]2 (2p3, which is the 706.722 nm line of the same group, so the printed label would appear twice).",
"evidence": "NIST's air wavelength of (2P*<3/2>).4p 2[3/2] J=2 - 4s[3/2]2 is 763.511 nm (within 0.0005 nm); 4p'[3/2]2 - 4s[3/2]2 is 706.722 nm."
},
{
"dataset": "A_4s_5p",
"row": "462.844",
"column": "x",
"printed": "4s'[1/2]1-5p'[5/2]2",
"corrected": "4s'[1/2]1-5p[5/2]2",
"reason": "A [5/2] level on the 2P1/2 core cannot exist: the level is 5p[5/2]2 (3p8), the level of Table I, printed primed in Table VI.",
"evidence": "The text and Table I's caption write 5p[5/2]2 unprimed (pp. 2462, 2464); the printed wavelength is the air wavelength of (2P*<3/2>).5p 2[5/2] J=2 to this lower level from NIST's levels (within 0.0003 nm) and Table I lists the same line."
},
{
"dataset": "A_4s_5p",
"row": "430.010",
"column": "x",
"printed": "4s[3/2]1-5p'[5/2]2",
"corrected": "4s[3/2]1-5p[5/2]2",
"reason": "A [5/2] level on the 2P1/2 core cannot exist: the level is 5p[5/2]2 (3p8), the level of Table I, printed primed in Table VI.",
"evidence": "The text and Table I's caption write 5p[5/2]2 unprimed (pp. 2462, 2464); the printed wavelength is the air wavelength of (2P*<3/2>).5p 2[5/2] J=2 to this lower level from NIST's levels (within 0.0003 nm) and Table I lists the same line."
},
{
"dataset": "A_4s_5p",
"row": "419.071",
"column": "x",
"printed": "4s[3/2]2-5p'[5/2]2",
"corrected": "4s[3/2]2-5p[5/2]2",
"reason": "A [5/2] level on the 2P1/2 core cannot exist: the level is 5p[5/2]2 (3p8), the level of Table I, printed primed in Table VI.",
"evidence": "The text and Table I's caption write 5p[5/2]2 unprimed (pp. 2462, 2464); the printed wavelength is the air wavelength of (2P*<3/2>).5p 2[5/2] J=2 to this lower level from NIST's levels (within 0.0003 nm) and Table I lists the same line."
},
{
"dataset": "A_4s_5p",
"row": "394.898",
"column": "x",
"printed": "4s[3/2]2-5p[1/2]1",
"corrected": "4s[3/2]2-5p'[1/2]1",
"reason": "The upper level is 5p'[1/2]1, not 5p[1/2]1 (the 425.118 nm line of the same group, so the printed label would appear twice).",
"evidence": "NIST's air wavelength of (2P*<1/2>).5p 2[1/2] J=1 - 4s[3/2]2 is 394.898 nm (within 0.0001 nm); 5p[1/2]1 - 4s[3/2]2 is 425.118 nm."
},
{
"dataset": "A_5p52_branches",
"row": "2831.415",
"column": "x",
"printed": "5s[5/2]2-5p[5/2]2",
"corrected": "5s[3/2]2-5p[5/2]2",
"reason": "The lower level is 5s[3/2]2: a 5s level has K = j_c = 3/2 or 1/2, never 5/2.",
"evidence": "Table I's energy, 113 468.55 cm-1, is NIST's (2P*<3/2>).5s 2[3/2]* J=2 (113 468.473 cm-1), and with 116 999.39 cm-1 gives the printed 2831.415 nm."
},
{
"dataset": "A_5p52_branches",
"row": "4938.586",
"column": "x",
"printed": "5s[1/2]1-5p[5/2]2",
"corrected": "5s'[1/2]1-5p[5/2]2",
"reason": "The lower level is 5s'[1/2]1 (the 2P1/2 core): K = 1/2 for an s electron needs j_c = 1/2.",
"evidence": "Table I's energy, 114 975.07 cm-1, is NIST's (2P*<1/2>).5s 2[1/2]* J=1 (114 975.019 cm-1), and with 116 999.39 cm-1 gives the printed 4938.586 nm."
}
]
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.