1 of 36

NIST Atomic Data, Databases and Tools �for Kilonova Studies

Yuri Ralchenko

Department of Astronomy, University of Maryland, College Park, MD 20742

Center for Research and Exploration in Space Science and Technology, NASA/GSFC, Greenbelt, MD 20771

X-Ray Astrophysics Laboratory, NASA/GSFC, Greenbelt, MD 20771

​

​

​

Coupling Radiative Transfer and Atomic Physics to Study r-Process Transients

14 Jan 2026, Los Alamos NM

Supported by NASA under award number 80GSFC24M0006

2 of 36

NIST ASG → NASA GSFC ASG

  • NIST ASG: terminated in May 2025
  • The group moved to NASA Goddard Space Flight Center (Greenbelt MD), formally employed by the University of Maryland College Park through CRESST II
  • Some experiments (VUV FTS, EBIT) are already moving to Goddard
  • All atomic databases will move to GSFC, so far accessible at the old NIST URLs; some hopefully will become accessible by this Fall, the rest by 2027

Yu. Ralchenko, 01/14/2026

3 of 36

Yu. Ralchenko, 01/14/2026

NIST/NASA High Resolution Atomic Spectroscopy

10.7-m normal incidence spectrometer

30 – 500 nm, λ/Δλ~1x105

2-m Fourier Transform Spectrometer

250 nm – 5.5 µm, λ/Δλ~4x106

  • Transition probabilities of neutral and singly ionized iron-group elements in the UV/VUV for abundance analysis and stellar atmospheric models

​

  • Calibration of iodine cells for radial velocity method of detecting exoplanets

​

  • Sandage-Loeb test using wavelength shifts in quasar spectra to measure fine structure constant over multiple decades

​

  • Calibration of ground and space- based spectrometers (Hubble Space Telescope)

Vacuum UV Fourier Transform Spectrometer

135-1150 nm, λ/Δλ~2x106

4 of 36

Yu. Ralchenko, 01/14/2026

VUV Fourier Transform Spectrometer

FT700

  • 140-600 nm using PMTs (longer into red with possible swap of detectors to IR-Diodes (not been done before on this instrument))
  • Minimum Resolution of 0.025 cm-1
  • Wavelength accuracy on the order of 108
  • Well known instrument response profile ideal for studying line shapes.
  • Linear intensity response of FTS is ideal for measuring Transition Probabilities of neutral, singly, and doubly ionized.

Sources

  • High-Current Hollow Cathode Lamp
    • Ideal for singly and doubly ionized spectra. HCL can produce higher ionization stages, usually with narrower line widths than Penning, but only for lower lying levels and for strong transitions.
  • Penning Lamp
    • Ideal for higher ionization stages (up to three-four times ionized) and better population of higher levels.
  • Combining both sources can usually produce spectra from neutral to triply-ionized.
  • Measurements of the spectra of rare-earth ions (Nd/Gd/Y) that can be produced in both sources (2-3 times ionized) typically use both sources. Data is selected from the source with better statistics.
  • Both lamps are tunable to better excite desired ionization stage and levels.

5 of 36

Yu. Ralchenko, 01/14/2026

First VUV FTS spectrum at GSFC: Fri Jan 9 2026

6 of 36

Yu. Ralchenko, 01/14/2026

NIST Atomic Databases and Tools

https://www.nist.gov/pml/atomic-spectroscopy-databases

7 of 36

Yu. Ralchenko, 01/14/2026

Workflow of critical evaluation of spectroscopic data

For the 16 energy levels of the 1s2l2l’ core-excited configurations of Li-like ions, 999 absolute and 35 relative experimental energy-level measurements from 101 publications have been collected and analyzed. Modern statistical theory methods (e.g., “dark uncertainty”) were applied to derive the final uncertainties.

A. Kramida

Fus. Sci. Tech. 63, 313 (2013)

8 of 36

Yu. Ralchenko, 01/14/2026

NIST ASD: example of C I (2017)

LEVELS

LINES

9 of 36

NIST Atomic Spectra Database v.5.12 (2024)

Yu. Ralchenko, 01/14/2026

https://www.nist.gov/pml/atomic-spectra-database-contents

~120,000 energy levels

~301,000 spectral lines

10 of 36

Yu. Ralchenko, 01/14/2026

Nuclear charge

Ion charge

NIST ASD v.5.12 (2024)

120,000 energy levels

NIST ASD v.5.12 (2024)

301,000 spectral lines

Sr

366 years to go!

11 of 36

Primary sources of data

  • Lanthanides
    • Martin, Zalubas, and Hagan, Nat. Stand. Ref. Data Ser., NSRDS-NBS 60, 422 pp. (Nat. Bur. Stand., U.S., 1978)
      • Many lines available but have not been added yet…
  • Actinides
    • Ac I-III: Kramida 2022
    • Th I-II: Redman et al, 2014
    • Mostly Reader et al,1980

Yu. Ralchenko, 01/14/2026

12 of 36

Yu. Ralchenko, 01/14/2026

La-Lu I-V

NIST Atomic Energy Levels and Spectra Bibliographic Database (Jan 2026)

Ac-Lr I-V

13 of 36

Yu. Ralchenko, 01/14/2026

​

La

Ce

Pr

Nd

Pm

Sm

Eu

Gd

Tb

Dy

Ho

Er

Tm

Yb

Lu

​

​

I

60

30

49

39

30

59

57

37

40

62

61

50

52

140

52

​

1-5

II

19

7

16

15

4

8

12

8

8

10

11

11

11

64

38

​

6-10

III

14

11

5

9

3

5

4

3

4

3

7

8

6

7

12

​

11-20

IV

9

12

10

14

6

8

6

6

6

6

6

16

12

10

6

​

21-50

V

4

6

7

9

2

4

3

2

2

2

2

3

4

7

4

​

51-65

NIST Atomic Energy Levels and Spectra Bibliographic Database

​

Publications on the first five ions of lanthanides and actinides 2010-2026

​

Ac

Th

Pa

U

Np

Pu

Am

Cm

Bk

Cf

Es

Fm

Md

No

Lr

​

​

I

37

49

27

48

23

25

22

21

20

24

24

22

19

31

22

​

1-5

II

8

19

3

12

3

5

2

2

2

2

3

3

4

5

8

​

6-10

III

12

15

5

7

3

2

3

2

2

2

2

2

2

2

4

​

11-20

IV

6

30

5

6

4

2

3

2

2

2

2

2

2

2

2

​

21-50

V

1

8

5

4

3

1

2

1

1

1

1

1

1

1

1

​

51-65

14 of 36

Ab initio atomic calculations: energy levels

  • Highly-charged (almost closed shell) ions: may be better than 0.0001%
    • Na-like ions (e.g., Os, Ir)
      • Nuclear charge radii can be determined from EUV spectra
    • Li-like dielectronic satellites from 1s2l2l’ (Yerokhin and Surzhykov, 2019)
      • Beats experimental data

​

  • Heavy neutrals and low-charged ions: a few per cent
    • Radžiūtė et al (2020)
      • Singly-ionized lanthanides Pr-Gd
      • GRASP2K MCDHF code
      • Pr II: 8%
      • Pm II: 12%
      • Sm II: 6%
      • Eu II: 8%
      • Gd II: 7%

​

Yu. Ralchenko, 01/14/2026

15 of 36

Example: Gd II

  • 321 energy levels
  • 465 spectral lines
    • 17 lines identified

​

Yu. Ralchenko, 01/14/2026

16 of 36

Yu. Ralchenko, 01/14/2026

https://physics.nist.gov/PhysRefData/ASD/lines_form.html

17 of 36

Example: Gd II

  • 321 energy levels
  • 465 spectral lines
    • 17 lines identified

​

Yu. Ralchenko, 01/14/2026

10,282 Ritz lines!

(all possible E1 lines)

18 of 36

Yu. Ralchenko, 01/14/2026

Lower level

​

Upper level

​

λ(theory)

λ(Ritz)

abs(%)

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_6D

9/2

3550

3664.14

3.12

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_6H

9/2

3634

3606.78

0.75

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_6H

11/2

3728

3707.41

0.56

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_6G

9/2

3755

3788.66

0.89

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_6H

13/2

3864

3878.9

0.38

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_6G

11/2

3884

3923.39

1.00

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_6G

13/2

4071

4145.08

1.79

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_8P

9/2

4097

4088.81

0.20

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_6F

9/2

4142

4117.56

0.59

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_6F

11/2

4302

4299.17

0.07

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_8H

9/2

4334

4219.41

2.72

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_8H

11/2

4417

4277.95

3.26

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_8H

13/2

4526

4382.57

3.27

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_8F

9/2

4763

4571.96

4.18

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_8F

11/2

4900

4677.24

4.75

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_8D

9/2

5264

5149.2

2.23

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_8F

13/2

5350

4999.0

7.02

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_8D

11/2

5418

5280.0

2.62

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_8G

9/2

5740

5487.5

4.60

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_8G

11/2

5967

5702.5

4.64

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.5d_8G

13/2

6139

5802.2

5.80

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.6s_6F

11/2

11531

11380.1

1.33

4f(7)8S0.5d_9D.6s_10D

11/2

4f(8)7F0.6s_8F

11/2

12933

12601.3

2.63

NIST dynamic Ritz wavelengths vs. Radžiūtė et al, ApJSS 248:17 (2020)

19 of 36

NIST-LANL Lanthanide/Actinide Opacity Database

Fontes et al, Mon. Not. Roy. Astron. Soc. 493, 4143 (2020)

  • LANL LTE opacities for 57 ≤ Z ≤ 70, 89 ≤ Z ≤ 102
  • Temperatures: 0.01—5 eV
  • Densities: 10-20—10-4 g/cm3
  • Photon energy grid: 14,900 points

Tables and graphs:

    • Total opacities
      • Scattering
      • Absorption
        • Bound-bound
        • Bound-free
        • Free-free

https://nlte.nist.gov/OPAC

K. Olsen, C.J.Fontes, C.L.Fryer, A.L. Hungerford, R.T. Wollaeger, O. Korobkin, Yu. Ralchenko

Fontes et al, Mon. Not. Roy. Astron. Soc. 519, 2862 (2023)

20 of 36

Yu. Ralchenko, 01/14/2026

NIST LIBS Database https://physics.nist.gov/PhysRefData/ASD/LIBS/libs-form.html

  • Calculation of Saha/LTE spectra for any element or combination thereof in ASD (optically thin)
  • Arbitrary plasma parameters (but Saha/LTE!) and spectrum resolution
  • ASD-based
  • Graphical services
  • Problem: not all ions, A’s

 

All needed data

may be in ASD

Low temperature, high density: Saha-LTE!

21 of 36

Yu. Ralchenko, 01/14/2026

Sm 50% + Gd 50%

22 of 36

Yu. Ralchenko, 01/14/2026

Sm 50% + Eu 50%

23 of 36

LTE vs NLTE

  • LTE
    • Atomic data: energy levels, statistical weights, radiative probabilities
      • Boltzmann + Saha
    • Maxwellian plasmas
    • Collisions >> radiation
  • NLTE
    • Atomic data: LTE plus collisional cross sections (rate coefficients), autoionization probabilities, CX cross sections, heavy particle cross sections…
    • Not necessarily Maxwellian plasmas
    • Collisional-radiative (CR) models/codes (case-dependent!)

Yu. Ralchenko, 01/14/2026

E.g., McWhirter’s criterion:

 

24 of 36

Yu. Ralchenko, 01/14/2026

Online Collisional-Radiative Code

  • FLYCHK was developed at LLNL (H.-K. Chung, R.W. Lee)
  • Time-dependent, non-Maxwellian, opacity effects, radiation field,…
  • Free registration
  • Primarily dense plasmas (Zn ≤ 79), ionization distributions, spectra, radiative power losses…
  • Superconfigurations…not the best for low-charge ions…

https://nlte.nist.gov/FLY

Argon, Te = 100 eV

Ne = 1e12, 1e14, 1e16, 1e18, 1e20, 1e22, 1e24

25 of 36

NLTE codes: how good are they?..

  • Verification
    • Rate equations are solved correctly
    • Analytical limits (LTE) reproduced
    • Convergence (e.g., with n) successful

​

  • Validation
    • Comparison with experimental spectra, line ratios, etc.
    • Check of the input atomic data
    • Check of missing physics (e.g., IP lowering or opacity)

​

Yu. Ralchenko, 01/14/2026

Chapter 8: Validation and Verification of

Collisional-Radiative Codes

26 of 36

Yu. Ralchenko, 01/14/2026

Meeting

Year

Location

Results

NLTE-1

1996

Gaithersburg, USA

Lee et al, JQSRT 58, 737 (1997)

NLTE-2

2001

Virtual Workshop

Bowen et al, JQSRT 81, 71 (2003)

NLTE-3

2003

Gaithersburg, USA

Bowen et al, JQSRT 99, 102 (2005)

NLTE-4

2005

Las Palmas de Gran Canaria, Spain

Rubiano et al, HEDP 3, 225 (2007)

NLTE-5

2007

Santa Fe, USA

Fontes et al, HEDP 5, 15 (2009)

NLTE-6

2009

Athens, Greece

 

NLTE-7

2011

Vienna, Austria

Chung et al, HEDP, 9, 645 (2013)

NLTE-8

2013

Santa Fe, USA

 

NLTE-9

2015

Paris, France

Piron et al, HEDP 23, 38 (2017)

NLTE-10

2017

San Diego, USA

Hansen et al, HEDP 35, 100693 (2020)

NLTE-11

2019

Las Palmas de Gran Canaria, Spain

 

NLTE-12

2023

Valladolid, Spain

​

NLTE-13

2025

Paris, France

​

NLTE Code Comparison Workshops

27 of 36

NLTE-13 (Sorbonne Univ, Paris, France)

  • 16 codes + variations
    • Many GBs of submitted data
  • National laboratories, universities, private businesses
    • USA, France, China, Spain, Israel, Japan
  • AI/ML session
  • 4 days of discussions and presentations
  • Steady-state cases
    • Cu, Kr, Xe
      • High energy density physics, ionization potential lowering, magnetic fusion
  • Time-dependent cases
    • C, Ar
      • Recombination from a bare ion, EUV lithography

Yu. Ralchenko, 01/14/2026

28 of 36

Yu. Ralchenko, 01/14/2026

Dozens of parameters to study

29 of 36

Yu. Ralchenko, 01/14/2026

Xe

​

Te = 3000 eV

​

ne = 1x1014 cm-3

Ionization distribution

30 of 36

Yu. Ralchenko, 01/14/2026

Z = 32 (Be-like Kr)

Ionization potential

Autoionizing states

Rescaled level populations

31 of 36

Yu. Ralchenko, 01/14/2026

Z = 32 (Be-like Kr)

32 of 36

Yu. Ralchenko, 01/14/2026

TD-C

From bare ion

Te = 1 eV

ne = 1018 cm-3

Mean ion charge

33 of 36

Yu. Ralchenko, 01/14/2026

Fractional PR rate

Population influx Xe28+

34 of 36

Acknowledgements

  • NASA/UMD
    • A. Kramida
    • J. Tan
    • J. Ward
    • G. Nave
    • K. Olsen
    • R. Ibacache
  • LANL
    • C.J. Fontes
    • C.L. Fryer

​

  • NLTE-13 participants

Yu. Ralchenko, 01/14/2026

35 of 36

Yu. Ralchenko, 01/14/2026

4-8 May 2026

Trieste, Italy

​

DEADLINE: Jan 25

�Topics

  • Modern methods for calculation of atomic structure, radiation, and collisions
  • Fundamentals of molecular spectroscopy and molecular processes in plasmas
  • Ionization distributions, spectral line intensities and collisional-radiative modeling
  • Spectroscopy of plasma-surface interactions
  • Diagnostics of high- and low-temperature plasmas in magnetic fusion devices
  • Astrophysical spectroscopy
  • Machine learning techniques for plasmas
  • Online atomic and molecular databases and tools
  • Data management and dissemination

​

https://indico.ictp.it/event/11140/

36 of 36

Conclusions

  • PLEASE tell us how we can help: elements, ions, spectral ranges, etc.
  • Experimental AS program at GSFC will include a significant component on lanthanides (and actinides?) spectroscopic measurements of wavelengths, energy levels, and transition probabilities
  • NASA atomic databases will be updated with the newly generated KN-related data; there are also some legacy datasets not in ASD – to be added soon
  • Validation and verification of NLTE codes is a must
    • Well developed and tested techniques exist

Yu. Ralchenko, 01/14/2026