Metallicity of stars

Astronomers call all elements more massive than Helium^3.5 `metals', and denote them by $Z$ ($X$ and $Y$ being the hydrogen and helium abundance by mass, respectively). These are produced in stars, but for some elements, like e.g. Carbon, we don't really know which stars are the dominant source. A lot of metals are produced during super novae (SNe) explosions. There are two basic types of SNe. SNe of type II are explosions of very massive stars. Nuclear fusion during the explosion is responsible for many elements of the $\alpha$-type, like Oxygen and Silicon. SNe of type I, on the other hand, are thought to occur during mass-transfer in binary stars - these are the dominant producers of Iron, Fe. The chemical mix of the stars is therefore a fossil relict of the relative fractions of type I and type II SNe. Typically though, one measures just one metallicity, for example the Fe abundance since that is often the easiest to measure, and then assumes that the other elements scale with Fe. But we know for a fact that some stellar systems have a rather different relative abundance pattern of the elements.

For the sun, the total amount of metals by mass, is about $Z_\odot=0.02$, or 2 per cent of the mass in the solar system is not hydrogen or helium. This is inferred not just from observations of the sun, but also from the composition of comets.

For other stars, one usually compares the metallicity in units of the solar value, on a logarithmic scale,

$$[Fe/H]\equiv log_{10}\left[(Fe/H)/(Fe/H)_\odot\right] .$$ (3.4)

So if a star has [Fe/H]=0, it has the same Iron abundance as the Sun, and for [Fe/H]=-1, it has one tenth the solar value.

One might expect the metallicity of a star to be related to its age. Indeed, very old stars formed before there had been many generations of stars^3.6, and hence before many SNe exploded, and so would have been formed from gas containing mostly hydrogen and helium. But stars forming now, will contract from gas that has already been polluted by SNe, and hence will be more metal rich. Within the MW, this seems to be born-out, at least to some extent. Stars in old GCs, for example, typically have [Fe/H]$\sim -1$, so quit a bit below the solar value. And most of the old stars in the halo also have low metallicity. These are called population II stars. In contrast, stars in the disk usually have higher metallicity, and are called population I. There is no strict divide between those, some disk stars also have low $Z$ for example.

Recently, there has been quite a bit of interest in the very first generation of stars that formed after the Big Bang. Those would have $Z=0$! They are called population III. The halo star with then lowest metallicity currently known, has [Fe/H] $\approx -5$, and might well be one of the first stars to form in the MW.

The evolution of $Z$ within the MW, or within galaxies in general, is called their chemical evolution (which is a misnomer, since the elements are produced in nuclear reactions which do not involve chemistry).

truecm

Table 3.1: Disks

truecm truecm

	Neutral Gas	Thin Disk	Thick Disk
$M/10^{10}\hbox{$M_\odot$}$	0.5	6	0.2 to 0.4
$L_B/10^{10}\hbox{$L_\odot$}$		1.8	0.02
$M/L_\B$ ($M_\odot$/$L_\odot$)		3
Diameter ( kpc)	50	50	50
Distribution	$\exp(-z/0.16{\hbox{\rm kpc}})$	$\exp(-z/0.325{\hbox{\rm kpc}})$	$\exp(-z/1.4{\hbox{\rm kpc}})$
$[Fe\/H]$	$>0.1$	-0.5 - 0.3	-1.6 - -0.4
Age (Gyr)	0 - 17	$<12$	14 - 17

truecm

truecm

Table 3.2: Spheroids

truecm truecm

	Central Bulge	Stellar Halo	Dark Matter Halo
$M/10^{10}\hbox{$M_\odot$}$	1	0.1	55
$L_B/10^{10}\hbox{$L_\odot$}$	0.3	0.1	0
$M/L_\B$ ($M_\odot$/$L_\odot$)	3	$\sim 1$	-
Diameter ( kpc)	2	100	$>200$
Distribution	bar	$r^{-3.5}$	$(a^2+r^2)^{-1}$
$[Fe\/H]$	-1 - 1	-4.5 - -0.5	-1.6 - -0.4
Age (Gyr)	10 - 17	14 - 17	17

truecm