Reference Label	Details
Gaydon 1968	A. G. Gaydon, "Dissociation Energies and Spectra of Diatomic Molecules", 3rd ed., Chapman and Hall: London 1968
Buttenbender 1935	G. Büttenbender and G. Herzberg, Ann. Phys. 413, 577-610 (1935) Über die Struktur der zweiten positiven Stickstoffgruppe und die Prädisssoziation des N2-Moleküls
note N2	CODATA gives 78715 +- 50 cm-1 and cites Buttenbender 1935 predissociation of C3Piu state of N2 as interpreted by Gaydon 1968. Gaydon 1968 quotes Buttenbender as 97970 +- 40 cm-1, and also gives Caroll 1965 as giving 97940 +- 40 cm-1. the value of Caroll 1965 appears on their limiting curve of dissociation, based on Buttenbender 1935. The dissociation goes to the N(4S)+N(2D) limit via avoided crossing. From NIST Atomic Web, the term values for the N(2D5/2) and N(2D3/2) are 19224.464 cm-1 and 19233.177 cm-1. The average of these two states is 19228.82 +- 4.36 cm-1. Assuming that the predissocation refers to the lower of the two (average in parentheses), Buttenbender 1935 value of 97970 +- 40 results in D0 of 78746 +- 40 cm-1 (78741 +- 40 cm-1), Caroll in 78716 +- 40 cm-1 (78711 +- 40 cm-1), and Gaydon 1968 quote of Buttenbender 1935 in 78746 +- 40 cm-1 (78741 +- 40 cm-1). Huber 1971 gives 9.7594 eV = 78715 cm-1 as arising from predissocation assuming dissociation into 2S3/2 and 2D5/2. Roncin 1984 reports from predissociation in c'4 1Sigmau+ state D0(2D+2D) for 14N2 as 117105.3 +- 22.7 and for 15N2 as 117167.0 +- 18.0 cm-1. They subtract 2*E(2D) of 38449 cm-1 from Moore NBS 34 (1970). In Moore NBS 35 (1971) we find 2D5/2 as 19223 and 2D3/2 as 19231 cm-1. However, NIST Atomic Web noted above gives 2D5/2 as 19224.464 cm-1, so 2*E2D5/2) = 38448.928 cm-1, which is essentially the value that was used by Roncin 1984. Hence, D0(14N2) = 78656.4 +- 22.7 cm-1 and D0(15N2) = 78718.1 +- 18.0 cm-1. They further correct these for simple anh. ZPE (no Y00, nor higher anh. terms, 15N2 from isotopic scaling) to get De, average the two that differ by 22 cm-1 and attach an uncertainty of +- 9 cm-1. Here we take D0(14N2) as such (with a tiny corr. to natN2 which increases D0 by 0.15 cm-1) producing 78656.6 +- 22.7 cm-1, and correct D0(15N2) to natN2 (-39.48 cm-1), giving 78678.6 +- 18.0 cm-1. The two actually differ by 22.0 cm-1. Lofthus 1977 quotes Buttenbender 1935 and Frackowiak 1964 as sources of data on predissociation of C3Piu of 14N2 and 15N2 (respectively). Lofthus gives a table of v and N dependent values of last unpredissociated and first predissociated levels for 14N2 and 15N2. From limiting curves of dissociation Lofthus 1977 obtain the limit above v=0 N=0 (F2) of 8960 +- 40 cm-1 for 14N2 and 9065 +- 45 for 15N2. The limit is assigned as (4S+2D). Lofthus 1977 quote T0(C = 88978 cm-1 for 14N2 and 88983 for 15N2 (calc. isotope shift). From these, they obtain D0(14N2 -> 4S+2D) = 97938 +- 40 cm-1 and D0(15N2 -> 4S+2D) = 98048 +- 45 cm-1. They use E(2D) = 19224 cm-1 to deduce D0(14N) = 78714 +- 40 cm-1 and D0(15N) = 78824 +- 45 cm-1, and claim (without specifics) that with the isotope shifts for the ZPE the two limits are consistent. With our isotopic correction from 15N2 to natN2, D0(15N2) becomes 78785 +- 45 cm-1, which seems barely consitent with D0(14N2)
as quoted by CODATA Key Vals	as quoted by CODATA Key Vals