Research by Regan W.
Introduction
There is a class of stars in the galaxy that are known to be metal-poor. Not much research has been done on the exact composition of these stars, but if we find out what heavy elements are present in them and in what concentrations, we can gain a better understanding of the early history of elemental production in the galaxy. The purpose of my research was to study which heavy elements were in metal-poor stars and to see if their abundances agree with current models developed by astronomers. (See Appendix for a more thorough explanation of nucleosynthesis processes)
Hypothesis
Do halo giant stars contain heavy elements? If so, what are they?
Comparing Two Stars
In this study I examined the ultraviolet spectra of two population II metal-poor stars: HD122563 is a classic metal-poor star of spectral type F/G. It’s located in the northern galactic halo in the constellation Bootes. Its chemical composition is similar to the sun, but its metal abundance is on the order of a hundred times less.
CS22892-052 is a star discovered by the Curtis Schmidt survey. It is a cooler star of spectra type K. It is located in the southern galactic halo in the constellation Aquarius. It’s heavy elements, such as rare earth elements, are reduced by a factor of a thousand compared to solar abundances.
The unusual chemistry of CS22892-052 indicates that there were different kinds of chemical processes happening in different parts of the galaxy early in its history.
Procedure
The first step of the process was to meet with Professor Charles Cowley of the University of Michigan and learn how to use the supplemental program called a double-precision synthesis program, in order to determine the presence of heavy elements. It would pull up a stellar spectrum and render tools for finding the position, line depth, and width for any given absorption feature.
With all that in place, a query to the Vienna Astrophysical Line Database (or the VALD) was made. The VALD is a collection of atomic line parameters of astronomical interest, which provides tools for selecting subsets of lines for typical astrophysical applications: Line identification, chemical composition and radial velocity measurements, model atmosphere calculations, etc.
A query to the Vienna Astrophysical Line Database (VALD) was made through the use of email. In order to get the query data, a star model had to be created. In this model, the "light" elemental abundances were reduced by a factor of a thousand, and the "heavies" abundances only down by a factor of a hundred. The VALD would create a synthetic spectrum for the portion queried, with all the lines identified, according to their known behavior.
The query starting point was placed in the ultraviolet range because little research has been done in that region. There are two reasons for this: spectra for the ultraviolet areas of stars do not have a well established database of lines for research and ultraviolet spectra is difficult to come by since almost data needs to be collected in outer space, outside of the earth’s atmosphere.
For each VALD query, the synthetic spectrum was compared to the actual spectrum. The features in the observed spectrum were matched to their corresponding lines in the virtual spectrum using a program called Matched Dates II. This program would extract the numbers from both the spectra and match up equivalent wavelengths. Match Dates was an application created by a radio astronomer written for a different use, but I figured it would work well in this project.
The line depths of the absorption features were calculated next. The lines depths are expressed in decimals, which reflect a percent of absorption in the spectrum. For example, 0.20 is equal to 20% absorption at that frequency in the spectrum. The VALD returned line depth data for every feature. To measure line depth in the actual star spectrum, the supplemental database was used to pull up five-angstrom portions of data for measurement on the computer. In order to maintain accuracy, a minimum line-depth deviation would have to be met. If these checks had passed, the VALD’s identification of the line could be assured as proof.
For both the stars (CS22 and HD122563) all lines that matched between the virtual and observed spectra were placed on a new spreadsheet, and then divided into two more spreadsheets. One spreadsheet lists matched lines with "standard" elemental identifications. These were lines caused by elements of atomic No. 30 and lower. The second spreadsheet lists matched lines with "heavy" elemental identifications (i.e., those lines caused by elements of atomic numbers higher than 30).
Differences in the line positions and depths for the "standard" and "heavy" matches were calculated next. This was done by subtraction.
Procedure (Part II)
Elemental Abundances
~ Heavy elements were selected that could be identified with high certainty due to identification over multiple wavelengths. The elements also had to have strong line depths in order for the abundance calculation to be accurate.
~ The double-precision synthesis program was used to determine abundances. They were calculated by altering the log of the oscillator strength value until the absorption feature of the synthetic spectrum matched the corresponding feature on the actual spectrum.
~ A new stellar model was piloted that calculated abundances by use of matching the strength of the H16 and H17 lines in the star. Previous studies calculated the abundances using the equivalent line width method.
Controls
The three points of controls for identifications:
Line Positions
Depths
Symmetry
If the lines had double-peaks, round tips, misshapen slopes, fat bottoms, or were asymmetrical at all, the VALD identifications could not be taken as absolute. These blends (features that are more than one line near the same position) add together to produce an asymmetrical feature. Blends were ruled out and taken as unidentifiable.
A wavelength match of less than 0.3 Angstroms and a line depth match of + or – 0.15 were used as cutoffs for determining valid line matches.
How To Read The Data Tables
*All Line matches are listed. Elements highlighted in pink were those that passed controls checks. *
VALD (WL)- The query wavelength given in angstroms.
ELEMENTAL ID- Element symbol and valence state.
VALD L.DEPTH- The query line depth.
STAR L.DEPTH- The actual star line depth that was calculated by subtracting the Cursor L.Depth from one.
CURSOR L.DEPTH- The star line depth recorded by the double-precision synthesis program.
L.DEPTH DIFF.- The difference between actual and queried line depths.
DEVIATION (MATCHED)- The difference between actual and queried wavelengths.
F/B- Indicate the direction of the deviation. F=Positive and B=Negative.
Click here to view data table: HD122563 Heavy Matched Lines
Click here to view sample database results: NPLTFTS graph
Analysis
CS22892-052 The spectrum from 3570-3696 A was studied. 71 standard elements and 49 heavy elements were identified after controls and error checking were applied. These heavy elements were identified by more than one line match, listed in order of decreasing certainty: W, Ru, Os, Mo, U, Nb, Th, Gd, Tm, Er, Nd, Zr, Rh, Dy.
HD122563 The spectrum from 2889-2936 A was studied. 22 standard elements and 3 heavy elements were identified after controls and error checking were applied. Only one heavy element was identified with multiple lines: Mo.
Comparing The Two Stars
HD122563 had fewer absorption features, which may be due to its lower surface temperature or the composition of its atmosphere. It may be due to one of two things: When this star formed in the galaxy, there weren’t as many of these heavy elements present.
Element Abundances
Abundances for three elements in CS22892-052 were calculated. Abundances are expressed as the element concentration relative to hydrogen plus all other elements.
Wavelength Element ID Abundance
3603.20 A Fe 3.10x10-08
3608.76 A Fe 6.98x10-08
3609.25 A Ni 1.64x10-08
3600.38 A Dy 1.36x10-12
3606.12 A Dy 8.58x10-13
3616.12 A Er 0.86x10-13
3613.12 A Dy 1.36x10-12
3620.20 A Dy 1.08x10-12
3621.20 A Sm 1.39x10-13
Average Dysprosium Abundance: 1.16x10-12
Average Iron Abundance: 5.1x10-08
* Abundances for the three heavy elements are plotted on the abundance graph above. *
Conclusions
Both of the halo giants stars were found to contain heavy elements. CS22892-052 contained more types of heavy elements and in greater abundance than HD122563.
The use of the hydrogen line fit model piloted in this study yielded different abundance values for some heavy elements than previously published studies using the equivalent line width fit model.
Graph of the abundances calculated for CS22892-052 showing their relative fits to the R and S nucleosynthesis processes.
This study by Chris Sneden et al. used the equivalent line width fit model.
Bibliography
Astrophysical Database: http://simbad.u-strasbg.fr
Astrophysical Database: http://vizier.u-strasbg.fr
Burris, Debra L et al. "Neutron-Capture Elements in the Early Galaxy: Insights from a Large Sample of Metal-poor Giants." The Astrophysical Journal, Vol. 544, Issue 1, 2000; pp.302-319.
Cowan, John J. et al. "The Thorium Chronometer in CS 22892-052: Estimates of the Age of the Galaxy." The Astrophysical Journal, Vol. 480, 1997; pp. 246.
Cowan, John. J. et al. "R-Process Abundances and Chronometers in Metal-poor Stars." The Astrophysical Journal, Vol. 521, 1999; pp. 194-205.
Jaschek, Carlos and Mercedes. The Classification of Stars. Cambridge University Press, Cambridge: 1987. (Reprinted with corrections 1990)
Sneden, Christopher et al. Evidence of Multiple R-Process Sites in the Early Galaxy: New Observations of CS 22892-052. The Astrophysical Journal, Vol.533, Issue 2: 2000; pp. L139-L142.
Sneden, Christopher et al. "The Ultra Metal-Poor, Neutron-Capture-Rich Giant Star CS 22892-052." The Astrophysical Journal, Vol. 467, 1996; pp. 819-840.
Sneden, Christopher et al. "Ultra metal-poor halo stars: The remarkable spectrum of CS 22892-052." The Astrophysical Journal, Vol. 431, 1994; pp. L27-L30.
Sparke, Linda S., and John S. Gallagher. Galaxies in the Universe: An Introduction.
Cambridge University Press: United Kingdom, 2000.
Vienna Astrophysical Line Database: www.tycho.astro.univie.ac.at/cgi-bin/vald