– 1 –1. Stellar Evolution For Massive Stars1.1. The Importance of Massive StarsIf one adopts a Salpeter IMF, there are 3 stars with M > 8M⊙for every 1000 starswith 0.1 < M < 120M⊙. Massive stars are very rare. But over short timescales they injectvery large amounts of radiation, mass, and mechanical energy into the ISM.Homology arguements are basically dimensional analysis of the equations of stellarstructure. For high mass stars they yield L ∝ M3and R ∝ M0.8, so Teff∝ M0.4. AlsoTc∝ M/R ∝ M0.2. The main sequence liftime τ(MS) ∝ M/L ∝ M−2, making mainsequence lifetimes of stars with M ∼ 100M⊙only about 300,000 yr.Thus a star with M = 100M⊙has L ∼ 106L⊙. Such a star can be detected as anindividual star beyond the local group even while on the main sequence.Although few in number, about 14% of the mass of stars in a generation is inM > 8M⊙. Almost all of this gets back into the ISM after stellar death; only a smallamount remains locked in compact remnants (neutron stars or BH); the maximum mass ofa neutron star is about 3 M⊙.Mass loss is important for massive stars. A Solar metallicity star withM(initial) = 60M⊙will lose 3/4 of its initial mass through stellar winds. Forv(wind) ∼ 3000 km/sec and dM/dt = 4 × 10−5M⊙/yr, t he mechanical luminosity is 30,000L⊙, or ∼10% of the star’s radiative luminosity, which is ∼ 603L⊙. Over a lifetime of500,000 years, this is a n energy input into the ISM of 2 × 1051ergs. So a massive star overits lifetime injects into the ISM about the same amont of energy as does a SN.– 2 –Details of massive star evolution determine:(a) the surface chemical composition of evolved massive stars, hence of nuclear processedmaterial into the ISM(b) the relative population o f Wolf-Rayet stars (which are losing mass very rapidly) versusnormal O type stars, and the distribution over the two types (WN vs WC) of WR stars.(The ratio n(WC)/n(WN) increases at higher metallicity.)(c) the ratio of red vs blue supergiants(d) the frequency of various types of core collapse SN (II - H lines present; Ib - no H lines,He lines present; Ic - no H or He f eatures in spectra).(e) the rotation rates of young pulsars1.2. Consequences of zero metalsa) Star formation is affected as the metals are normally important in providing coolingduring the collapse of a gas cloud into a proto-star. Details discussed in section on starformation.b) lower opacities, especially in the UV, will tend to raise the luminosity of a massive 0Zstar sinceL(rad) = 4πR2[4π3κRρacT3dTdr].c) Without any CNO for catylizing nuclear reactions, the CNO H-burning process cannotrun. A 0Z massive star must burn H using the much slower pp chain. Since this is a slowprocess, but the stellar luminosity is high, the central temperature must be very highTc∼ 108K, to produce the required energy. To achieve such a high Tc, the star mustcontract, using gravitational energy to cover for its deficit of nuclear energy. The star– 3 –continues gravitational contraction to a smaller ra dius than would occur for a star with thesame mass but with Solar Z. This continues until enough12C is produced by burning Hefrom the Big Bang via the 3α process, which will run at this high Tc, to ignite the CNOcycle of H burning.d) With regard to stellar evolution, the lower mean ato mic weight due to absence of heavyelements is insigificant. The somewhat lower He abund (Y ∼ 0.24 vs 0.28) is not veryimportant.e) In late stages o f stellar evolution when there is mass loss, there will be no dust in theouter layers of the stellar atmosphere and the stellar wind. Molecules will have to form inthe gas phase, not on the surfaces of dust grains. Radiatively driven winds will not be asimportant as at higher metallicity, as it is the radiation pressure on dust (for luminous coolstars) and the line emission in a tomic lines (for luminous hot stars) that normally drivessuch winds.f) Ignoring the effect of rotatio n, given their lower mass loss rates, massive stars will endtheir lifetimes at higher final mass than a solar metallicity star of the same initial mass.Pair instability SN, which leave no remnant, ejecting all the star’s mass into the ISM, willbe more frequent at 0Z.g) With regard to the chemical evolution of the ISM, in the Solar neighborhood we observethat ∆Y / ∆Z ∼ 2. At very low metallicity, Y remains a lmost constant at the Big Bangvalue as Z steadily increases. The small increment in Y from stellar evolution over severalstellar generations is very small compared to the initial Big Bang He abundance. This doesnot hold for heavier elements not produced in the Big Bang , hence not present at all atZ = 0.The combination of an extended contraction phase and the low opacity leads to 0Z stars– 4 –that are more compact, with smaller radii, than solar Z stars. A 0Z 20M⊙star has aradius 3.5 times smaller than that of star with the same mass at Solar Z. A smaller radiusleads to a shorter timescale to mix from center to surface for a given value of the diffusioncoefficient, and to steeper gradients in angular momentum.1.3. Detecting 0Z Massive StarsNo 0Z star has ever been confirmed thro ugh spectroscopy in our galaxy, in its halo ,or in any of the nearby g alaxies for which such studies are possible. The Teffof 0Z starswith M > 90M⊙is predicted to be approximately constant at 105K, while that of Solarmetallicity stars in the same mass range is almost constant at 50,000 K. The higher Teffof a 0Z star of a given mass than a MS star of the same mass with Solar metallicity, bothof which have approximately the same total luminosity, means the former will emit manymore ionizing photons in the UV. The 0Z massive star will ionize a much la rger numberof H atoms in the surrounding ISM, and produce very strong recombination lines, muchstronger than for a higher metallicity star. The harder radiation field will produce intenseHeII (1640˚A) emission.A search for the intense HII regions expected from massive 0 Z stars a t high redshiftmay be possible with JWST or with the TMT. The SKA should be able to image neutral Hat high redshift and track the reionization of the IGM with time. LOFAR, a proposed largearray of low frequency radio detectors, is designed to carry out tomogrpahy of the redshiftedHI out to z ∼ 30, when the 21 cm spin-flip HI transition is shifted to 50 MHz. This projectfaces a very large sky background which will make any detection very challenging.The fact that