PDA

View Full Version : H/He abundances at early epochs



quanta
2004-Oct-24, 12:29 PM
How are the relative abundances of hydrogen, helium and lithium actually detected. What wavelengths are observed that allows a determination of these abundances, particularly at early epochs.
I suspect that the 21cm hydrogen wavelength is prevalent in our galaxy and at low redshift, but from the little I know about this, it seems that only the higher energy photons (Lyman series etc) are detected at early epochs.

So many questions :)

Cougar
2004-Oct-24, 03:39 PM
How are the relative abundances of hydrogen, helium and lithium actually detected. What wavelengths are observed that allows a determination of these abundances, particularly at early epochs.

This information might be somewhat dated since Nobelist Steven Weinberg's The First Three Minutes, A Modern View of the Origin of the Universe was originally written in 1977. But in it he says....

Estimates of the primordial helium abundance of the universe are based on comparisons of detailed calculations of stellar evolution with statistical analyses of observed stellar properties, plus direct observation of helium lines in the spectra of hot stars and interstellar material.

Regarding deuterium....


Our knowledge of the cosmic deuterium abundance was put on a much firmer basis by ultraviolet observations in 1973 from the artificial Earth satellite Copernicus. Deuterium atoms, like hydrogen atoms, can absorb ultraviolet light at certain distinct wavelengths, corresponding to transitions in which the atom is excited from the state of lowest energy to one of the higher states. These wavelengths depend slightly on the mass of the atomic nucleus, so the ultraviolet spectrum of a star whose light passes to us through an interstellar mixture of hydrogen and deuterium will be crossed with a number of dark absorption lines, each split into two components, one from hydrogen and one from deuterium. The relative darkness of any pair of absorption line components then immediately gives the relative abundance of hydrogen and deuterium in the interstellar cloud.

Normandy6644
2004-Oct-24, 04:24 PM
I need to read that book. I read "The Final Theory" and thought it was awesome. I also like how he can write nontechnical books like those and also technical books, with both is gravitation and cosmology book as well as his volumes on field theory being standard texts.

ngc3314
2004-Oct-24, 07:14 PM
How are the relative abundances of hydrogen, helium and lithium actually detected. What wavelengths are observed that allows a determination of these abundances, particularly at early epochs.
I suspect that the 21cm hydrogen wavelength is prevalent in our galaxy and at low redshift, but from the little I know about this, it seems that only the higher energy photons (Lyman series etc) are detected at early epochs.

So many questions :)

At this point, we have lots of data from high redshift on the occurrence of Lyman absorption from intergalactic gas, and for z>2.8 or so, ionized helium through the analogous helium lines (He II Lyman alpha is at 1216/4=203 A, Lyman limit at 288 Angstroms emitted). However, in both cases there is good reason to think that we are seeing only a minority ionization state of the material, so getting absolute abundances takes additional information. Ratios among elements are a bit easier to deal with, if you pick pairs that have similar ionization potentials. Within galaxies, we can use emission-line ratios for gas ionized by hot stars or active nuclei. For star-forming galaxies, we find that Oxygen in particular is low (but not enormously so) around z=3 (while quasars at high redshift may in fact have higher abundances of certain metals, which makes sense only if they are in very particular dense star-forming enironments and catch supernova debris). Recently, we've been able to compare far-UV spectra of stellar populations in local and high-z galaxies, showing that some elements (like phosphorus, which should come from red giants and take a long time to appear) are particularly low.

Eta C
2004-Oct-24, 08:15 PM
The Particle Data Group review on Big Bang nucleosynthesis (http://www-pdg.lbl.gov/2004/reviews/bigbangnucrpp.pdf) discusses this. To quote the site (sect 20.2) "One seeks astrophysical sites with low metal abundances, in order to measure light element abundances which are closer to primordial." This is not an easy task as systematic errors need to be dealt with. Still, primordial abundances of deuterium, helium 3 and 4, and lithium 7 have been measured. The review gives the various references for the measurements.

quanta
2004-Oct-24, 10:24 PM
Thanks for the replies.
Interestingly, it confirms my suspicions that only the high energy spectrum (Lyman UV) of these elements seems to be the prime candidate for observational abundance at high redshift. Anyone seen the 21 cm hydrogen wavelengths at early epochs?
Also, highly ionized iron X-ray emissions seem to dominate the observed spectrum at these epochs.
Don't we observe the lower energy photons at these redshifts (or is that self explanatory by virtue of their low energy)?

ngc3314
2004-Oct-24, 11:07 PM
Thanks for the replies.
Interestingly, it confirms my suspicions that only the high energy spectrum (Lyman UV) of these elements seems to be the prime candidate for observational abundance at high redshift. Anyone seen the 21 cm hydrogen wavelengths at early epochs?
Also, highly ionized iron X-ray emissions seem to dominate the observed spectrum at these epochs.
Don't we observe the lower energy photons at these redshifts (or is that self explanatory by virtue of their low energy)?

There are ony a few 21cm absorption (or emission) detections, mostly because receivers typically haven't worked well at those wavelengths (except when the H I line redshift carried it across a longer band already in use). There is one galaxy in front of quasar at z=0.5 or so with H I seen (more information at http://www.astr.ua.edu/keel/agn/3c196.html). As new facilities specifically designed for lower frequencies come into use (LOFAR, Square Kilometer Array), a major objective is seeing 21 cm radiation from the early Universe. Eventually, after the epoch of the CMB, we expect to see a ripply sea of 21cm emission spanning redshifts down to about 10. The problems in removing background and foreeground radiation are formidable; we expect to see the ripples but not necessarily measure how strong the mean level is with much accuracy.

HIghly ionized X-ray lines show up only in high-energy source - quasars and galaxy clusters. Some of these lines get easier to see with redshift just because the redshift carries their emission into bands where better detectors are available.

The redshifted UV lines are powerful tracers for two reasons - they will be by far the strongest spectral features of H and He unless they are mostly ionized, and they are redshifted into the optical range where spectroscopy is much cheaper than other wavelength bands.

quanta
2004-Oct-25, 02:14 AM
from ngc3314

The redshifted UV lines are powerful tracers for two reasons - they will be by far the strongest spectral features of H and He unless they are mostly ionized, and they are redshifted into the optical range where spectroscopy is much cheaper than other wavelength bands

Thats interesting and makes sense. So its more of a practical consideration, but have we actually detected Balmer series (as an example) from hydrogen at high redshift, or is there no real effort made to detect anything other than the wavelengths which are 'cost effective'

ngc3314
2004-Oct-25, 02:40 AM
from ngc3314

The redshifted UV lines are powerful tracers for two reasons - they will be by far the strongest spectral features of H and He unless they are mostly ionized, and they are redshifted into the optical range where spectroscopy is much cheaper than other wavelength bands

Thats interesting and makes sense. So its more of a practical consideration, but have we actually detected Balmer series (as an example) from hydrogen at high redshift, or is there no real effort made to detect anything other than the wavelengths which are 'cost effective'

There are some measurements of Balmer lines at z>3 by now. The problems are instrumental and atmospheric - atmospheric absorption takes big bites of the spectrum, and thermal emission from the atmosphere eats into sensitivity. There have now been major samples, both with Keck, Subaru, and especially the ESO VLT, covering various pieces of the redshifted visible spectral. (Hubble can do this for wavelengths shortward of about 1.7 microns with NICMOS, but at much reduced wavelength resolution). This lets us use the same line ratios ([O III]/H-beta, [N II]/H-alpha) commonly used for emission-line abundances nearby. The Balmer lines are also seen in high-z quasar spectra, showing up into the IR as well. Now NGST ought to do this very well indeed...

One sample (from a page on a collaboration a couple of years old) of seeing [O III] emission so strong that H-beta is only a limit can be seen at http://www.astr.ua.edu/keel/vlt , from the superb ISAAC instrument on the VLT. This is a galaxy at redshift close to z=3. This also shows some of the difficulties of sky subtracton and absorpion - the noise level changes dramatically over even short wavelength spans.

quanta
2004-Oct-25, 11:16 AM
Thanks ngc3314 (and others). Thats just the information I was after.
I just started reading your referenced site- Very impressive :)