JOURNAL OF RESEARCH of the National Bureau of Standards-A. Physics and Chemistry The Second Spectrum of Nickel (Ni II) (Addenda and Errata) A. G. Shenstone* Institute for Basic Standards, National Bureau of Standards, Washington, D.C. 20234 (April 12, 1971) This paper provides additional information which came to light too late for the original publication [J. Res. Nat. Bur. Stand. (U.S.) 74A, 801 (1970)]. Key words: Energy levels; ionization potential; nickel; spectral series; spectroscopy; wavelength. 4. The levels of w1D° are doubtful because each of them depends on only two lines, one of which is doubly assigned. The levels y "G31/2 and y 4G412 may be spurious and x F1, is definitely so and should be discarded. The line list includes all the lines which were assigned to this level. 11/2 5. Two new levels whi h I assign to 3d7 (2P)sp(1P) as w2P1/2 (135053.14) and 2D (136461.10) have been found. They are companions to known doublet levels. The analysis of Ni II is by no means complete, but it could not have been carried even as far as it has been without the essential aid of Professor Shadmi. That is not to say that we are in complete agreement on all of the allocations of levels to the calculated terms. Our failures to agree arise, I think, from the fact that the theorist's chief concern is the numerical fit whereas the experimentalist is interested in other criteria, specifically in this case the intensities of combinations. I believe my allocations are more realistic because most levels do show rather definite characters when the intensities are examined. I shall quote only two cases. In the group of levels 3d8(3F)5p there are three levels with J=42. Their total intensities of combination with quartet and doublet levels are approximately as follows: 1500 to 15 for 104081, 2000 to 15 for 104298 and 45 to 1000 for 105588. Professor Shadmi chooses the first of these as 2G41/2 and the last as 4G41/2" a choice no experimentalist would ever make. A second example is my x 4G31/2, for which Shadmi's assignment is 'D31/2 The intensities of combination with the 42, 32, 2/2 of s 2 4F are 2, 8, 25 which is perfect for 4G31/2 and could not be worse for 1D31/2 6. Table II, page 843, fourteen Observed Levels 52205.95 to 78955.45 inclusive, the "ConfigurationTM column should be corrected to read 3d 4s2. 7. Table II, page 845, thirteen Observed Levels 128924.94 to 128966.52 inclusive, the "Configuration" column should be corrected to read 3d8(3F4)5g. 8. Table II, page 846, three Observed Levels 134334.46 to 134336.68 inclusive, the "Configuration" column should be corrected to read 3d8(3F4)6g. 9. Table III, page 849, the name of the Observed Level 118877.09 should read 4f 'Pi/, instead of 4f 'P12 10. Table III, page 852, the name of the level 135746.06 should read v 2G1, instead of o 2G1/2" There are four levels which Shadmi has been unable to fit at all into the scheme. The level v 2D212 at 135258.88 is undoubtedly real, making one combination with a quartet level and four with doublet levels. The term v4 P which fails to fit by such very large amounts appears to be real from the experimental evidence. I hope that in the future I can continue to get such great help from my friends in the Bureau of Standards and Israel as I have had during the past years when I was engaged in the analysis of Ni II. (Paper 75A4-672) JOURNAL OF RESEARC+ the Vamanat Sureau of Standards - A Physics and Chemistry Tables of Second Virial Coefficients and Their First and Second John S. Gallagher and Max Klein Institute for Basic Standards, National Bureau of Standards, Washington DC 20234 (April 22, 1971) Expressions are newbet ir the second virial coefficient and to feat two tem je Key words Dies: terniecular potential function, polar, second 1. Introduction Effects on the thermodynamic properties of nonspherical molecules dae to the presence of angular dependent terms in the potential finetion can be quite important. In some cases, these efects cause large deviations from two parameter corresponding states. Such angular dependent terms can be placed in two categories. There are those which are due to “permanent" orientation dependent fieres resulting from the shape of the isolated molecules. In addition, there are induced nonspherical forces which arise from the change in the shape of a molecule due to the proximity of a second molecule. The former can be quite large while the latter are generally small. A significant dif ference between the two types of terms is associated with the fact that the relevant coeficients in the former (e.g., the dipole moment are, in principle, usually measureable in experiments based on isolated atom effects (e.g. spectrocopy, see for example [1]') while those for the latter can rarely be determined separate and so are most often investigated with the help model and a parameter. Effects due to permane orientation dependent forces are thus much useful in the study of such effects on thermodymusee properties both because they are large and man they do not involve additional parameterization The formalism for including orientational both the permanent and induced variety in frontali tial function has been developed by severa [2. 3. 4. 5. 6. 7]. In the case of molecules with gern nent dipole moments. Stokmaver accounted for. *Supported in part i ment Center. T.. F Figures in bra 1. Arnow & .. where r*=ro, a= 6 m -6 1 )( m 6 172 6 X=2 cos 0; cos 02 + sin 0, sin 02 cos (Þ; +Þ2), fitting of experimental data for polar substances and = 1-(cos 201 + cos 26 with the angles defined as in figure 1, where u is t The choice of the (m, 6, 3) potential function family dipole moment and a is the mean polarizability of t was, of course, somewhat arbitrary. A study of the sensitivity of thermodynamic properties to the potential function in the case of spherical molecules (i.e. nonpolar in the present context) showed all reasonable three parameter functions to be equivalent when it came to predicting such properties [9]. Subsequently a very successful new correlation of nonpolar second virial coefficients was produced in which the authors arbitrarily chose to base their correlation entirely on the (m, 6) function [13]. Furthermore, tables of collision integrals have now been published for the spherical (m, 6) potential function [14]. The spherical (m, 6) function has also been extended to quantum fluids with the preparation of tables of Wigner-Kirkwood corrections to the second virial coefficient for the (m, 6) family [15]. It thus seemed appropriate to select the (m, 6) as the form for the central part of the potential. With the publication of the present tables. there be comes available an extensive set of tables for the study of various variations on the (m, 6) potential model. e FIGURE 1. molecule, assumed to be isotropic. It should be note The nonspherical terms in (1) have a strong effe on the potential function. Rowlinson [12] has plotte the potential (1) for m = 12 for different dipole strength including in the potential only the spherical and dipol terms, the latter for dipoles in the end-on position. H results show that the actual well depth for this end-o configuration for a reduced dipole moment 7(= μ2/€σ of 4.0 is eight times the well depth for the nonpola (12, 6) potential. In addition, he found that the pola part of the potential has the effect of making the side of the potential well much steeper. In particular Rowlinson found that the ratio of the coordinate at th potential minimum to that at the potential zero (i.e Tmin, changed from 1.414 for 7--0 to 1.002 for 7=4.0 These strong modifications of the spherical part o the potential function result directly from the inclusion of the effect of the dipole moment and so indicate strong sensitivity of the predicted second virial coeff cient to the magnitude of the dipole moment used. |