Energy levels of the second spectrum of curium, Cm I
Earl F. Worden
Lawrence Livermore National Laboratory, University of California, Livermore, California 94550
John G. Conway
Lawrence Berkeley Laboratory, University of California, Berkeley, California 94720
Jean Blaise
Laboratoire Aim& Cotton, Centre National de la Recherche Scientifique, 91405 Orsay, France
Received December 9, 1985; accepted January 27, 1986
The curium emission spectrum from electrodeless lamps has been observed from 2400 to 26 500 A. About 30% of the14,250 observed lines have been assigned by experimental methods as Cm II lines. Analysis of the spectrum hasproduced 432 odd levels and 162 even levels of Cm II that combine to classify more than 4,100 curium lines. The en-ergy levels are listed, and most have Land6 g values from Zeeman effect data as well as isotope shift values. Manylevels have been assigned to electronic configurations and terms by use of isotope shift and Lande g values. A tablecontaining the lowest level found for each of the eight identified electronic configurations is given. The groundstate of the ion is the 8S/ 2 level of the (Rn) 5f 77s2 configuration. The next electronic configuration is 5f 87s, whoselowest level is the 8F13/2 level at 2093.87 cm-'.
INTRODUCTION
In 1976 we reported on the energy levels of the first spectrumof curium, Cm .1 References to spectroscopic work on curi-um up to 1975 are given in that paper. The earliest papers2 '3
gave wavelengths for about 200 lines and 244Cm-242Cm iso-tope shifts for 183 lines. In a second paper in 1976 wereported wavelengths, intensities, and classifications (whenknown) for 2034 of the strongest of more than 13,250 curiumlines emitted by an electrodeless discharge lamp (EDL) andobserved photographically on large grating spectrographs inthe 2450-11 500-A region.4 The curium lines observed inthe infrared by a Fourier-transform spectrometer were alsoreported in 1976.5 Wave numbers and intensities of 1734lines between 8400 and 26 500 A were listed, and classifica-tion of 87% of the lines was given. A number of new energylevels of Cm I were found.
The energy levels of the Cm II spectrum tabulated here arebased on these reported observations and on extensive un-published isotope shift studies6 (see the section headed Ex-periment). The isotope shift data together with previouslymeasured Zeeman effect data permitted confirmation of anumber of levels that were found through searches for con-stant sums and differences. They also allowed assignmentof these levels to electronic configurations with considerableconfidence. As a result, levels belonging to the odd configu-rations 5f
77S
2, 5f7 6d7s, 5f7 6d2 , and 5f87p and to the even
configurations 5f87s, 5f 86d, 5f7 7s7p, and 5f76d7p were iden-tified. A total of 432 odd and 162 even levels were found,and 94 of these were assigned to the listed odd configura-tions and 100 to the listed even configurations. An abstractof this work has been published.7 Preliminary results weregiven in abstracts, 8 9 and an interpretation of the 5f8 7s con-figuration of Cm II was given in a generalized parametricstudy of fh and fs configurations in the actinides.10
Other published work on the emission spectrum of curiumis by the Russian investigators at the I. V. Kurchatov Insti-tute of Atomic Energy. 1",2 One" reported 6771 lines be-tween 2424 and 7009 A emitted by EDL's and observed withlarge grating spectrographs. Iron lines were used as stan-dards both as iron impurity in the curium lamps and as ironlines photographed adjacent to the curium exposures.Lines of other impurities such as Li, Na, Mg, Al, Ca, Ti, Zn,Cu, and Am were identified and eliminated from the list byRussian investigators, but we have found that their list stillcontains more than 800 impurity lines belonging to Pu, Ba,La, Ce, Pr, Nd, Sm, Eu, and Gd. A number of lines wereindicated as neutral or singly ionized based on their relativeintensities from lamps operated at different microwave fre-quencies (90 and 300 MHz) and power levels. They alsoindicate that self-reversal was observed in 320 lines. Thesecond paper' 2 reported isotope shifts for 97 lines as ob-served from EDL's with three even isotopes (80.3% 244Cm,17% 2 4 6 Cm, 1% 2 4 8 Cm) and two odd isotopes (1% 2 4 5 Cm and0.7% 247Cm). A number of 246Cm-244Cm shifts were ob-served in the 2430-6932-A region, but Lobikov et al. report-ed shifts only for the 97 lines where all three even isotopepositions were measured.'2 Of these, 38 were measured withan interferometer and are accurate to 0.01 cm-'. The re-maining 59 were determined with a large grating spectro-graph.
EXPERIMENT
SourcesElectrodeless discharge lamps were used exclusively for allour observations of the curium spectrum. The techniquesused for preparation of the lamps and employing them toobtain wavelengths, Zeeman effect, spectrum assignment,
I I I I tlit If lilIlilh hl JJJJJJ l ~~~~~II I 1 Ili 1111 111111 IIIL
I 111111 1111 I 1 cm 1 g = 1
Fig. 1. Zeeman patterns of one Cm I and two Cm II linesphotographed on the ANL 9.15-m spectrograph.
EnergyLevel J(cm-') Odd
reversal,
ir
H = T
at 2.4 T
self-reversal, and isotope shift have been discussed."5 ,'3"14
The isotopic composition of the curium samples used toprepare the EDL's is given in Table 1. The type of spectro-scopic data obtained with lamps prepared from samples isgiven in the "Use" column of the table. Approximately 100to 200 ,ug of curium was used for each lamp preparation.Less than 2 mg total material was used in the investigation,and most of the material was recovered for other uses.
WavelengthsThe method of recording the curium spectra has been de-scribed fully in Ref. 1. The spectra were photographed onthe Argonne National Laboratory (ANL) 9.15-m Paschen-Runge spectrograph in the wavelength region 2400 to 9100A. Additional spectra were photographed on a 3.4-m Ebertspectrograph by using a grating with 300 lines/mm at anglesnear 590 (2400-3200 A) and 300 (8900-11 500 A). Linepositions were measured on semiautomatic comparatorswith an accuracy of +1 4m. The photographic wavelengthlist contained about 13 250 lines due to 244Cm. The wave-number precision or internal consistency from the level anal-ysis is 0.02 cm'1 or better throughout the line list. Thenear-infrared region was investigated by using a Fourier-
Details of spectrum assignment procedures are given inRefs. 1 and 14.
Zeeman EffectThe Zeeman effect was photographed from 2400 to 9000 Awith the ANL 9.15-m spectrograph at a field of 2.4 T (24,000G). Figure 1 shows Zeeman patterns obtained for two Cm IIlines. The patterns were measured with a semiautomaticcomparator and the data reduced by computerized least-squares methods.1 The accuracy of the g values derivedfrom resolved Zeeman patterns such as 3962.7 A in Fig. 1 is+0.003 Lorentz units. Level g values obtained from suchmeasurements are given to three significant figures after thedecimal point in Tables 2 and 3. Level g values reported totwo places after the decimal are accurate to 0.02 Lorentzunits. These values were derived from measurements onunresolved patterns or on patterns where only a measure ofthe interval between adjacent components (delta g) could beobtained. A more detailed description of how g values werederived is given in Ref. 1.
Isotope ShiftSpectra of curium isotopes obtained by using lamps withisotopic compositions given in Table 1 were recorded on the9.15 m ANL spectrograph and were measured with a semiau-tomatic comparator. Figure 2 is a sample portion of thespectrum obtained with these isotopic mixtures. The ad-vantage of using lamps with three isotopic compositions isevident from this figure. Shifts of more than 6000 lines weremeasured. The shifts are the wave number of the heavyisotope minus the wave number of the light isotope. Shifts
Isotope(percent)
248 (55) v246 (18) °244 (27) 0
244 (95) 0
246 ( 3) 0
N~ CV)
o 0
V. V Vo 0 0
* 0 a
V00
0
0
q
transform spectrometer. A total of 1734 lines were mea-sured with an accuracy of 0.005 cm-1 .5 Because of the over-lap, the infrared measurements added about 1000 lines tothe list.
Spectrum AssignmentWhen operated at low metal-atom pressures, EDL's tend toenhance the second spectrum, and at high metal-atom pres-sures the first spectrum is predominant. In the region 2400to 9400 A, 51% of the lines were assigned to Cm I and 30% toCm II, while the remaining 19% were mostly weak lines.
244 (55) 245 (23) x246 (21)
*0 0 0 0x x xo o 0 0 0
1 cm- 1
Fig. 2. 3A region of the curium spectrum showing the isotope shiftfor four isotopes. Spectra of three isotopically different samples areshown. Photographed on the ANL 9.15-m spectrograph.
Worden et al.
Vol. 3, No. 8/August 1986/J. Opt. Soc. Am. B 1099
242 244 246 248
.10 ± 0.07 1.000 0.954 ± 0.012
+ 0.012
245Fig. 3. Relative isotope shift of five isotopes of curium from atomic emission spectra.
14
12
E
0
I-
w
10
8
6
4
2
0
5/2 9/2 13/2 1/2 5/2 9/2 13/2
iFig. 4. Low-lying energy levels of singly ionized curium. The electronic configurations and LS-term designations are given.
of clearly resolved lines, for example the three lines with
wavelength identified in Fig. 2, have a precision of +0.005cm'1 or better. Level shifts obtained from these lines arereported to three significant figures in Tables 2 and 3. Level
shifts reported in the tables to two places are precise to
about +0.02 cm-'. The isotope shift of the ground state wasassumed to be zero. Lines with shifts less than 0.1 cm- 1 areprecise to only about +0.01 cm'; a few shifts less than 0.1cm'1 are precise to only about ±0.05 cm-'. Line isotopeshifts less than 0.1 cm-l were obtained from values mea-
sured for 248 Cm-244 Cm shifts. They were converted to24 6Cm-244Cm shifts by use of the relative isotope shift (RIS)
factor of 1.945 for (248Cm-244 Cm)/(2 46 Cm-2 44Cm) derived
from lines that have large isotope shifts and are unblended.The RIS for 245Cm was measured by using lines with small orno 245Cm hyperfine structure. 245Cm has a nuclear spin of
7/2. These Cm RIS values are shown in Fig. 3 together witha RIS for 242Cm derived from 244Cm-242Cm shift data of
Conway and McLaughlin. 3
RESULTS
The energy levels of singly ionized curium atoms are listed inTable 2 (odd levels) and Table 3 (even levels). Column 1
Worden et al.
1100 J. Opt. Soc. Am. B/Vol. 3, No. 8/August 1986
+0.2 - ODD
x
-0.2 -
-0.4
-0.6
-0.8 -
1.0 _
1.2 1-
Fig. 5. Range of observed isotope shifts for several electronic con-figurations of Cm II. The value of x is about +1.2 cm- 1 (see text).
Table 4. Lowest Levels of Identifiedin 244Cm II
Configurations
Energy IsotopeLevel g Shift Configuration(cm-') Value (cm') Designation
lists the value of the energy level in inverse centimeters.Columns 2 and 3 give the J and g values. Column 4 is theisotope shift in units of inverse centimeters. The isotopeshifts are 246Cm-244Cm shifts relative to the 5f
77s
2 s7/2ground state as zero shift. Column 5 gives the configurationdesignation followed by the term, usually in LS-couplingnotation. In some cases the Ji-coupling notation is used.We followed the configuration notation used in Atomic En-ergy Levels-The Rare-Earth Elements.'5
Figure 4 shows the level structure of the lowest terms ofCm II. The electronic configurations and LS designationsare indicated. The level structure of the 8F and 6F terms ofthe 5f87s configuration found at the beginning of Table 3 is
more clearly illustrated in this figure. Figure 5 shows therange of isotope shifts for levels assigned to identified con-figurations of Cm II. Note that there is overlap of observedshift for a number of configurations. In these cases, theory,Land6 g values, transition intensities, and other factors areimportant for assigning the level to the appropriate configu-ration. As seen in Fig. 5, we report the shifts relative to theground state as x. For the shifts in Tables 2 and 3, the X iszero. The value of x is approximately +1.2 cm'1 for Cm II.In heavy elements the largest isotope shifts are found forlevels belonging to configurations with the largest number ofpenetrating s electrons and the smallest number of shieldingf or d electrons. The smallest isotope shifts are for levels ofconfigurations with no penetrating electrons. Thus, in thecase of Cm II, the configuration 5f 86d is expected to have ashift close to zero. As can be seen in Table 4, this leads to avalue of approximately +1.2 cm' for levels of 5f7752. Thelevels of the still-unidentified 5 66d7s2 configuration areexpected to have larger shifts.
Table 4 lists the lowest level of the identified configura-tions of Cm II along with their designations, g values, andisotope shifts. We have identified levels belonging to all theconfigurations expected below 50 000 cm-' except the 5f9configuration predicted to start around 26 000 cm-', accord-ing to Brewer.' 6
A parametric study of only the 5f 87s configuration 0 wasmade before the more complete isotope shift data were avail-able. As a result, several levels identified as 5f87s werereassigned to other configurations or were removed. At thesame time, new levels with the correct isotope shift wereassigned to 5f 87s. These corrected assignments are includ-ed in Table 3. The levels of 5f 87s below 18 000 cm-' did notchange. Since these levels strongly influence the fit, theparameters in Ref. 10 will not be changed significantly.
ACKNOWLEDGMENTS
We wish to thank Mark Fred (ANL) for the use of spectro-graphic facilities and computer programs. Thanks also toE. K. Hulet (Lawrence Livermore National Laboratory) forsupplying the curium samples and to G. V. Shalimoff and T.Parsons (Lawrence Berkeley Laboratory) and R. Burns(ANL) for purification of the isotopically enriched samples.We thank R. G. Gutmacher for help with some of the experi-mental work and M. Jepson for keypunching many of theexperimental data. We appreciate the assistance of theHazard Control personnel, especially P. Linnes, in handlingand transportation of the radioactive materials. We wish tothank J. Verges [Laboratoire Aime Cotton (LAC)] for takingthe infrared spectra at LAC and J. F. Wyart (LAC) forcalculations that were valuable guides in the search for newlevels. J. Blaise wishes to acknowledge a grant from NATOthat made possible a visit to the Lawrence Berkeley Labora-tory.
This work was supported by the U.S. Department of Ener-gy under contract W-7405-ENG-48 at the Lawrence Liver-more National Laboratory; by the Director, Office of EnergyResearch, Office of Basic Energy Sciences, Chemical Sci-ences Division, under contract DE-AC03-76SF00098 at theLawrence Berkeley Laboratory; and by Centre National dela Recherche Scientifique at Laboratoire Aim6 Cotton, Or-say, France.
E
Q
C)
CD
N
0.04-'0UM
Cm I
EVEN
5f 6
Worden et al.
W 6d7s
Vol. 3, No. 8/August 1986/J. Opt. Soc. Am. B 1101
REFERENCES
1. E. F. Worden and J. G. Conway, "Energy levels of the firstspectrum of curium, Cm I," J. Opt. Soc. Am. 66, 109-121 (1976).
2. J. G. Conway, M. F. Moore, and W. W. T. Crane, "The emissionspectrum of curium," J. Am. Chem. Soc. 73, 1308-1309 (1956).
3. J. G. Conway and R. D. McLaughlin, "Isotope shift and wave-lengths of the curium spectrum," J. Opt. Soc. Am. 46, 91-93(1956).
4. E. F. Worden, E. K. Hulet, R. G. Gutmacher, and J. G. Conway,"The emission spectrum of curium," Atom. Data Nucl. DataTables 18, 459-495 (1976).
5. J. G. Conway, J. Blaise, and J. Verges, "The i.r. spectrum ofcurium-244," Spectrochim. Acta 31B, 31-47 (1976).
6. E. F. Worden and J. G. Conway, "Isotope shifts in the atomicspectrum of curium," J. Opt. Soc. Am. 70, 1576A (1980).
7. J. Blaise, J. Verges, J. F. Wyart, J. G. Conway, and E. F. Worden,"Term analysis of the spectrum of singly ionized curium(Cm II)," presented at the European Conference on Atomic Phys-ics, April 6-10, 1981, Heidelberg, Federal Republic of Germany.
8. E. F. Worden and J. G. Conway, "Status of the analysis of thecurium I and II spectra," Physica 33, 274 (1967).
9. E. F. Worden and J. G. Conway, "Extension of the analysis of
Cm I and II spectra," presented at the Atomic SpectroscopySymposium, U.S. National Bureau of Standards, Gaithersburg,Md., September 11-14, 1967.
10. J. Blaise, J. F. Wyart, J. G. Conway, and E. F. Worden, "Gener-alized parametric study of the Sf and 5fp7s configurations,"Phys. Scr. 22, 224-230 (1980).
11. E. A. Lobikov, N. K. Odintsova, and A. R. Striganov, "Emissionspectrum of curium," I. V. Kurchatov Institute of Atomic Ener-gy, Rep. IAE-3210, Moscow (1979); translated by G. V. Shali-moff, Lawrence Berkeley Laboratory, Berkeley, Calif. 94720(personal communication, 1980).
12. E. A. Lobikov, A. R. Striganov, V. P. Labozin, N. K. Odintsova,and V. F. Pomytkin, "Atomic spectrum of curium," Opt. Spec-trosc. (USSR) 46, 596-599 (1979) [Opt. Spektrosk. 46,1054-1060 (1979)].
13. F. S. Tomkins and M. Fred, "Electrodeless discharge tubescontaining rare earth and heavy element halides," J. Opt. Soc.Am. 47, 1087-1091 (1957).
14. E. F. Worden, R. G. Gutmacher, and J. G. Conway, "Use ofelectrodeless discharge lamps in the analysis of atomic spectra,"Appl. Opt. 2, 707-713 (1963).
15. W. C. Martin, R. Zalubas, and L. Hagan, Atomic Energy Le-vels-The Rare-Earth Elements, NSRDS-NBS 60 (U.S. Gov-ernment Printing Office, Washington, D.C., 1978).
16. L. Brewer, "Energies of the electronic configurations of thesingly, doubly, and triply ionized lanthanides and actinides," J.Opt. Soc. Am. 61, 1666-1682 (1971).
