Astronomy Without A Telescope – Stellar Archaeology

[/caption]

Although, as we look further and deeper into the sky, we are always looking into the past – there are other ways of gaining information about the universe’s ancient history. Low mass, low metal stars may be remnants of the early universe and carry valuable information about the environment of that early universe.

The logic of stellar archaeology involves tracking generations of stars back to the very first stars seen in our universe. Stars born in recent eras, say within the last five or six billion years, we call Population I stars – which includes our Sun. These stars were born from an interstellar medium (i.e. gas clouds etc) that had been seeded by the death throes of a previous generation of stars we call Population II stars.

Population II stars were born from an interstellar medium that existed maybe 12 or 13 billion years ago – and which had been seeded by the death throes of Population III stars, the first stars ever seen in our universe.

And when I say death throes seeding the interstellar medium this includes average sized stars blowing off a planetary nebula at the end of their red giant phase – or bigger stars exploding as supernovae.

So for example, the low metal spectral signature of HE 0107-5240 matches that predicted for a very early low mass Population II star built from the end-products of a Population III supernova.

This is about as close as we can get gathering any information about Population III stars. Telescopes that can look deeper into space (and hence look further back in time) may eventually spot one – but it’s unlikely that any still exist. Theory has it that Population III stars formed from a homogenous interstellar medium of hydrogen and helium. The homogeneity of this medium meant that any stars that formed were all massive – in the order of hundreds of solar masses.

Stars of this scale, not only have short life spans but explode with such a force that the star literally blows itself to bits as a ‘pair-instability’ supernova – leaving no remnant neutron star or black hole behind. Supernova SN2006gy was probably a pair-instability supernova – mimicking the last gasps of Population III stars that lived more than 13 billion years ago.

Recipe for a pair instability supernova. In very massive stars, gamma rays radiating from the core become so energetic that they can undergo pair production after interaction with a nucleus. Essentially, the gamma ray creates a paired particle and antiparticle (commonly an electron and a positron). The loss of radiation pressure as gamma rays convert to particles results in gravitational collapse of the star's core - and kaboom! Credit: chandra.harvard.edu

It was only after Population III stars had seeded the interstellar medium with heavier elements that fine structure cooling resulted in disruption of thermal equilibrium and fragmentation of gas clouds – enabling smaller, and hence longer lived, Population II stars to be born.

Around the Milky Way, we can find very old Population II stars in orbiting dwarf galaxies. These stars are also common in the galactic halo and in globular clusters. However, in ‘the guts’ of the galaxy we find lots of young Population I stars.

This all leads to the view that the Milky Way is a gravitational hub nearly as old as the universe itself – which has been steadily growing in size and keeping itself looking young by maintaining a steady diet of ancient dwarf galaxies – which, deprived of such a diet, have remained largely unchanged since their formation in the early universe.

Further reading:

A. Frebel. Stellar Archaeology – Exploring the Universe with Metal-Poor Stars http://arxiv4.library.cornell.edu/abs/1006.2419

61 Replies to “Astronomy Without A Telescope – Stellar Archaeology”

  1. Seems logical to me to assume that Population 3 stars likely remain as the (Black holes) in galactic centers.

    I’m interested what comes next in stellar evolution. As more complex heavy materials form in stellar and planetary systems, so new stars forming now should be even more robust and long lived. (continuing trend?)

    How does a hypothetical Population 0 star look like? How will it differ from what we see today. Do new stars continue to form increasingly heavier elements. ? Will we ever get unobtainium forged in stellar cores? hehe.

    And where are we going to go with this counterintuitive way of classifying stellar bodies once we do have a population 0 star. 🙂

    Jokes aside, Stellar archeology needs to consider where stellar evolution is going as well. We are hamstrung by this natty old speed of light limitation, we can only look to the past and our present is still thousands or millions of years in our future.

    I mean before the light of new stars reaches us.

    Damian

  2. I don’t know. We may get a totally different archaeology of universe once we are able to consider the dark matter into the picture. But for now it’s a good story.

  3. @ damian

    Seems logical to me to assume that Population 3 stars likely remain as the (Black holes) in galactic centers.

    As the text says, Pop3 stars blow as pair-instability Supernovae implying that there is NOTHING left behind. Thus, they cannot form the super massive black holes (SMBH) of galactic centers. SMBH form differently, however, nobody knows exactly how. Accretion will definitely play a role, but how the “seed black hole” forms in the first place is unknown. Still, they are not remnants of Pop 3 stars.

    I’m interested what comes next in stellar evolution. As more complex heavy materials form in stellar and planetary systems, so new stars forming now should be even more robust and long lived. (continuing trend?)

    Yes and no. At the moment and also for a quite long part of the future you are right. This is due to the fact that metals (elements heavier than helium) boost convection in stars. This means that hydrogen can be replenished in the core (stars like the sun use only about 10% of their hydrogen for fusion), which means that the star can last longer in the hydrogen burning phase, which is the most efficient and, thus, longest phase in the life of a star.
    However, since more and more hydrogen of the interstellar medium is fused into heavier elements, less and less hydrogen is abundant for new stars. This means that in a distant future the hydrogen burning phase will also be very short or will cease entirely, since there is not much hydrogen available any more.

    How does a hypothetical Population 0 star look like? How will it differ from what we see today.

    Maybe those stars that form almost entirely without hydrogen….

    Do new stars continue to form increasingly heavier elements. ? Will we ever get unobtainium forged in stellar cores? hehe.

    No. Stars can only fuse elements up to iron, since fusion above iron would COST the star energy instead of gaining it. To say it technically: Fusion above iron is endothermic. No star will ever do it 😉 . But this leads to a stable iron core in stars which will ultimately cause its doom, once the iron core becomes to massive: KABOOM! A neutron star or a black hole is formed that way.

    And where are we going to go with this counterintuitive way of classifying stellar bodies once we do have a population 0 star. 🙂

    😀

  4. Damn, I forgot something:

    Heavier elements than iron are formed in Supernova explosions, where enough energy is available to combine them! (Yes, the Supernova loses energy that way!)

  5. Those mediocre pop-III stars weighing in at below about 250Msun [limit uncertain] are expected to disrupt completely (leaving no remnant).

    Any heavier ones are expected to collapse entirely as a result of the same pair-instability, forming a black hole suitable for seeding a SMBH. Whether heavier stars can form or not is still uncertain.

    Although the lighter ones does not immediately form a black hole, the effect they have on the surrounding gas-reservoir, aswell as the ‘polluting’ effect, probably means another star in the vicinity will form a black hole in relatively short timescales. If nothing else, the star should have developed a shell-type gas-reservoir during its lifetime, with possible secondary starformations triggered either during the lifetime, or by the explosion.

  6. This article mostly involves stellar astrophysics, which I really have a “101” knowledge of. But some thoughts come to mind with this. I will start with some issues of how matter got clumped into things like popIII stars and galaxies and where dark matter fits into this.

    The occurrence of galaxies is due to small anisotropies in the early universe. The universe went through a period of exponential expansion that was driven by a vacuum energy that was 10^{123} times that which exists today. This is a de Sitter spacetime configuration. The universe experienced some 50-60 so called e-folds, or its radius expanded by a factor e^n, for n = 50 or 60. This is a substantial expansion, which occurred rather rapidly in the first 10^{-20} seconds of the universe. Now this vacuum energy was driven by some type of scalar field, usually called the inflaton field. As this expansion occurred the wavelength of the field was stretched out and for many Fourier modes it was stretched out beyond the cosmological horizon. Back then the cosmological horizon was around the size of a proton, which is very small compared to today where it is about 10 billion light years. This is a reflection of how the vacuum state of the inflationary universe was huge, with the mass density of the entire universe packed into a volume in a few cubic meters. As this scalar field was stretched out beyond the horizon length it ceased to causally propagate and become locked into place, which includes quantum fluctuations of the field.

    Then this period ended, where the vacuum energy plummeted to near zero or the value we have today. The reasons for this involve the nature of a rather odd potential for this scalar field, which I will leave to anyone to look up. This locked field density plummeted as well, but not all at the same time. As a result pockets of field fluctuation entered this reheating period at different times. This then seeded the anisotropy of the universe, which is seen in the CMB and in the occurrence of galaxies. These local regions of clumped matter-fields included both ordinary luminous matter and dark matter. Then the universe went through this period of intense radiation domination, until the surface of last scatter which we see as the CMB. From there luminous matter clumped and dissipated energy and began to form these popIII stars within dark matter halos.

    PopIII stars were odd balls in some ways. Hydrogen is not very opaque and so they appear to have been somewhat transient implosions of matter that entered a rapid fusion period and then exploded into supernova. At this stage we leave cosmology and enter astrophysics. I can’t say much about the explicit nature of models for popIII stars. However, it appears that what happens is that the core becomes so hot that you begin to get pair production of electrons and positron as X-ray photons scatter with protons. This has an actual cooling effect as energy goes into more degrees of freedom and raises the effective heat capacity of the core. Curiously this mimics the physical configuration or phase of the early universe after post-inflationary reheating and prior to the time we see on the CMB. Since this cools the core some it leads to a further implosion of the core and increases the pressure. This can then accelerate the fusion rate, causing more pair production and … , which runs away into this supernova.

    Why this did not form early black holes is a bit of a mystery to me. This is where my lack of knowledge of stellar astrophysics leaves me hanging. It would seem to me that the runaway process could produce a shock wave inwards and outwards, where the inwards shocking implodes a part of the core into a black hole. It is my understanding that large black holes appeared on the scene pretty early on. Further, black hole were not generated in the above inflationary and post-inflationary period — or at least not many were produced. This would increase the entropy of the early universe in ways not possible. So it is during this so called reionization phase, where popIII stars occurred, and when large black holes likely emerged. So if popIII supernova did not generate black holes, then what did? Dark matter did not implode to form black holes, for it is too weakly interacting to dissipate energy sufficiently. These black holes did not emerge from the earliest universe during inflation and reheating. They might have emerged during the radiation dominated phase, but there are problems with this scenario. A soup of radiation and highly relativistic particles can’t increase gravitational clumping efficiently.

    So we have maybe a bit of a question here.

    LC

  7. I understand the ‘Lambda cold dark matter theory’ is the favored model of early large scale structure formation,
    see: http://en.wikipedia.org/wiki/Lambda-CDM_model

    and hence agree it’s vital to consider dark matter in any discussion of early galaxy formation. It’s less clear what the role of DM is in first star formation.

    I understand it’s difficult to explain why stars would form out of a homogenous gas cloud in the first place – albeit once they get started they should then grow to huge sizes – and once they start exploding there’s lots of turbulence to drive the formation of others.

    The puzzle is why particles first start to accrete from that initial dust cloud to form objects big enough to generate gravitational attraction. So maybe DM has a role in seeding this process somehow.

  8. The gravitation clumping was set up by the anisotropy induced by inflation and its end at reheating. Here the inflaton field enters the reheating phase at differential locations where these amplitudes become amplified. This results in clumps that involve both dark matter and luminous or ordinary matter. Now let us run forwards to the end of the radiation dominated phase into the matter dominated phase. At this phase atoms of luminous matter in these clumps interacted with each other by gravitation, but by other interactions which are much stronger. These are the electromagnetic and nuclear interactions, and so some extent the weak interactions. These involve particle collisions which release bosons, in the case of electromagnetism these are photons. Electromagnetism is the easiest to consider, which tells us collisions between charged particles results in the production of photons that largely leave the system. So the system loses energy, which results in a collective collapsing. Dark matter does not do this as it is very weakly interacting. Newtonian gravity predicts a recurrence of motion, but with ordinary matter we have other strong forces, which remove degrees of freedom from the gravitational motion of particles, which removes energy from the system. The result is collapse, which then results in the formation of these popIII stars.

    Somebody asked about the next generation stars. I would suspect that as time goes on stars will become more dwarfish. The current popI stars are already largely dwarf stars, and G-class stars represent only some 5% of stellar population. I would suspect that if one came back in 10 billion years this galaxy, or the merged outcome of the Milky Way and Andromeda galaxy, will largely consist of dwarf stars with very few G class stars, and further on in time there will be only M class stars, and then in 100 million years there will be few stars at all. Things are winding down, the second law of thermodynamics is marching everything onwards to cosmological heat death.

    LC

  9. I am just a little shocked and dismayed with the article here and the responses. I could do my usual tirade in blitzing the contributions, but I have really held off for a little while to see what people were thinking.

    In summary, much of the problem here is with theoretical cosmologists trying to describe stellar evolution, which is mostly beyond their own field of expertise, and this has lead I think with the misunderstandings here.

    Pop III stars are actually different than what we might call “first stars”, because Pop III reflected only the chemical composition, and henceforth, questions their evolution or more precisely evolutionary differences.

    “First stars” maybe vastly different from what we perceive as stars today – whose composition and behaviours were influenced by density and even residual anti-matter or with the concentration of dark matter.

    Firstly the names are Pop I, II and III for these stars and not 1, 2 or 3. Originally the divisions were with population of stars in galaxies. The difference is based solely on their spectra and the lines we can observe. Hence, say, a solar like star that is Pop I will show different line strengths than the young Pop II star whose lines are more prevalent. From this was established as means of determining the age and behaviour between the population types. (Not the differences here are not by age but by chemical composition. Our galaxy is really a blend of Pop I and II stars, and are not really clearly separate in regions of the galaxy. [Unlike, say, as Wikipedia wrongly suggests.]

    In 1944 Walter Baade (1893-1960) first realised that the nature of these Pop I and II stars suggest they might be born at different times when the Galaxy was formed. Old population stars formed first, while those on the spiral arms for after this. Chemical differences not only are different in the stars we see but were different when they were formed as the so-called Zero-Aged Main Sequence or ZAMS. Furthermore, the initial composition were divided into the XYZ parameter, being basically Hydrogen (X), Helium (Y) and all the other elements heavier than Helium (known also as ‘metals’ as the Z parameter (Z). Most ZAMS for newborn Pop II stars today start off as X=75% Y=23%-24%, and Z=1%-2%, but for Pop I stars the balance finds Z=0.02% to 0.2%. For this reason it is postulated that the enrichment of metals is caused by the redistribution of nucleosynthesis products by mass loss that once generated in the stars. It is also assumed that in the early universe was only 75% hydrogen and about 23-25% helium and crucially no metals. Hence the stars that formed in the early universe where purer mixtures gases than today’s stars – making the theorised Population III stars.

    Now stellar evolution models have shown that the stars evolve different with different levels of metals. This effects the range of stellar masses (especially with the upper range), the rate the star ages, and even the behaviour of the star.

    Astronomers, as they have the last thirty years, have been looking for Pop III stars in our galaxy, which are mostly on the form of white dwarfs. You would know a Pop III white dwarf from its spectral signature, which would appear as almost pure hydrogen and helium; actually being the dense hot atmosphere of the white dwarf. (There are less metals because the hydrogen only converts to helium.)

    This become the general definition of a Pop III star.

  10. The article says;

    “And when I say death throes seeding the interstellar medium this includes average sized stars blowing off a planetary nebula at the end of their red giant phase – or bigger stars exploding as supernovae.”

    Sorry this is quite wrong.

    Planetary nebulae do not “blow off” material. The mass loss, and the majority of the mass loss, is caused during the AGB phase (Asymptotic Giant Branch) – when the star is rising dramatically in luminosity and size. It is the hotter core of the star that causes this gross swelling where the outer atmosphere density makes it easily blown into interstellar space – either by ordinary winds or superwinds. 70% of the mass of the star is lost during the AGB phase, at a rate determined by Riemer’s Formula;

    Mass Loss = 10^-13 x Luminosity x Radius / Mass

    For stars less than 2.25 solar masses this is about 10^-10 solar masses per year. Stars above 2.25 Solar Masses a maximum this is about 0.0001 (10^-4) Solar mass per year.

    Planetary Nebulae expose the mass loss in the last 50000 years or so, where the UV energy illuminates the nebulosity.

    Therefore, the mass loss (enrichment) of the interstellar medium occurs during the whole period of the red giant phase. Furthermore, much of the material is mostly hydrogen and helium, with only a small percentage being ‘metals.’

  11. The article says;

    “Theory has it that Population 3 stars formed from a homogenous interstellar medium of hydrogen and helium. The homogeneity of this medium meant that any stars that formed were all massive – in the order of hundreds of solar masses.”

    This is also quite wrong.

    Frankly, these stars would actually be smaller if the interstellar medium was homogenous. The rate that the material being gathered is clearly dependant on the collapsing mechanism and how the forming disk is making some newborn star. It is far more likely more of the material during the collapse would make many stars of small mass (binaries or multiple stars) than one single big star.

    Another possibilty that is speculated is that stars bigger than 260 solar masses could collapse to a black hole completely missing the ‘star’ stage! (These are not really “Pop III” stars, but are the slightly different “First Born” stars, as eluded previously.)

    However, regardless of these interesting ideas, the question of the elimination of angular momentum remains the problem (just like star formation for the stars today.)

    As a comment, the idea of larger mass for Pop III was originally postualted to be due to the chemical make-up not the medium which it comes from!

    I do suggest interested readers here might like to read the Stacy et.al. 2009 paper The first stars: formation of binaries and small multiple systems,; especially with the “Introduction” and the “Summary and Discussion” sections.

    (The reference article in the one above, that appears in Science; Turk M. J., Abel T., O’Shea B., Sci., 325, 601 (2009) is excellent on this subject here.)

  12. Hon. Salacious B. Crumb,

    I wrote the bit about cosmology to indicate the conditions that may have existed that generated these first stars. As I say there has to be some measure of inhomogeneous distribution of matter for the onset of these first stars. I also admitted that my understanding of the astrophysics of popIII stars is pretty minimal.

    However, this does present a challenge for observational astronomy. As yet no popIII stars have been found, from my understanding. Yet if they are found at large z, say z ~ 10, at the reionization period we might be able to better understand these early conditions of the universe. The goal is to find if there is some matching between the astrophysics of these stars and cosmology.

    LC

  13. One of the other concerning issues here is the perceived assumption that all the Pop III stars were wiped out as supernovae. This I truly doubt, mostly because of the huge issues remaining in the eliminating angular momentum during their formation.

    IMO in the referenced article says under the first paragraph; 2.1 The First Stars, which states;

    “According to cosmological simulations that are based on the cold dark matter model of hierarchical structure growth in the Universe, the first stars formed in small minihalos some few hundred million years after the Big Bang. Due to the lack of cooling agents in the primordial gas, significant fragmentation was largely suppressed so that these first objects were very massive (of the order to ~100 Solar Masses; e.g, Bromm & Larson 2004 and references therein). This is in contrast to low-mass stars dominating today’s mass function. These objects are referred to as Population III (Pop III) as they formed from metal-free gas.”

    Is unlikely this is totally true, as stellar evolution constraints suggest smaller stars also existed with high-massed stars – as suggested in more than 100s of papers.

    The paper poses the question in the beginning of the paper, “What is the nature of Pop III stars?”, but totally ignores the possibility that the majority of these stars might just be mostly of smaller mass! [After reading this whole paper, I think it is narrowly talking about “Pop III Supernovae” rather than necessarily ALL the Pop III stars.]

    Again, Pop I, II and III stars are not distinct groups of stars, but through a whole range of metallicities. As the article alludes, metallicities are expressed in terms of Z or as a chemical ratio I.e. [Fe/H] – Iron to Hydrogen. [This same idea comes from with Beers, T.C., Christlieb, N., Annual.Review A&A., 43, 531 (2005) (as referenced by Frebel too) Sadly there is no copy to freely download on the net, hence no link to it.]

    Frankly, the definition of Pop III stars needs to be reassessed because the cosmologist’s concept seems to be radically different from that of the stellar evolutionist. This is mostly because they are looking at the problem from two different angles. I.e. Cosmology for supernova in high redshift galaxies, while stellar evolution is searching for stellar white dwarf embers of the first stars in the Milky Way.

    If we do assumed ALL the Pop III destroyed themselves as Pop III supernova, then there would be absolutely zero remaining Pop III white dwarfs in the Milky Way (and the stellar evolution astronomers are just wasting their time!)

  14. @ Hon SBC

    There’s a big difference between declaring something wrong and not agreeing with it.

    I can provide reference citations for various statements you have fired salvos at. This is not an area with a lot of scientific consensus. If you want that, please email me.

    Your claim that AGB mass loss is somehow a seperate process to planetary nebulae formation conflicts with the general literature, see:
    http://en.wikipedia.org/wiki/Planetary_nebula (first sentence)

    I acknowledge that my preference for 3 instead of III (etc) is atypical. My research for this article certainly left me with the impression that “there would be absolutely zero remaining Pop 3 white dwarfs in the Milky Way” (if the theory is correct).

    Stellar arch has it that old Pop 2 stars offer the best chance of assessing the composition of the interstellar medium that the Pop 3 stars left behind.

  15. “Your claim that AGB mass loss is somehow a separate process to planetary nebulae formation conflicts with the general literature, http://en.wikipedia.org/wiki/Planetary_nebula (first sentence)’

    No. I did not say it was a separate process. It is the end of the process, when the PNN (Planetary Nebula Nucleus) is exposed – the actual white dwarf of the star. Once the core is exposed there is no more significant gas to dissipate. Saying “…stars blowing off a planetary nebula” is wrong, because most of the gas being dispersed was from the earlier AGB phase. I.e. If the winds are blowing off at 1000 km per sec, and the planetary nebula is 0.2 parsecs across, how does the ejected material to travel the intervening distance? Actually, for a 2.0 Solar Mass star half of this mass (1.0 solar mass) of that has already been lost into interstellar space and is too far away to be illuminated by the central star when it is finally exposed!

    A general standard reference for this is Iben Jr’s; The life and times of an intermediate mass star – In isolation/in a close binary http://adsabs.harvard.edu/abs/1985QJRAS..26….1I

    The relevant section is 4.1 Mass loss (pg.19.), especially;

    “From the standpoint of its influence over the final fate of an intermediate mass star, an overwhelmingly important feature of the AGB phase of evolution is mass loss from its surface.”

    Iben also says (pg.24);

    “When a star leaves the AGB as a consequence of superwind mass ejection, the mass of its remaining hydrogen-rich envelope is of the order of one-tenth of the mass….

    This infers that most of the mass loss occurs during the AGB phase. As Fig. 16 (pg.25) points out;

    “The evolutionary track might be appropriate for a star of initial mass about 2 Solar Masses which loses about 1.2 Solar Masses during the AGB phase via an ordinary wind and then ejects a nebulae shell of mass about 0.2 Solar Masses after developing a C-O core of mass about 0.6 Solar Masses.”

    The illumination of the Planetary Nebula occurs after the ejection of the shell when the exposed core can UV fluoresce the expanding shell. Hence, most of the ejection has occurred BEFORE the appearance of the planetary nebula.

    NOTE: As for quoting the Wikipedia article is quite extraordinary. I actually wrote this originally for this Planetary Nebula article, and I even modified this exact same sentence on 9th May 2010!

  16. “Your claim that AGB mass loss is somehow a separate process to planetary nebulae formation conflicts with the general literature, [en.wikipedia.org Planetarynebula] (first sentence)’

    No. I did not say it was a separate process. It is the end of the process, when the PNN (Planetary Nebula Nucleus) is exposed – the actual white dwarf of the star. Once the core is exposed there is no more significant gas to dissipate. Saying “…stars blowing off a planetary nebula” is plainly wrong, because most of the gas being dispersed was from the earlier AGB phase. I.e. If the winds are blowing off at 1000 km per sec, and the planetary nebula is 0.2 parsecs across, how does the ejected material to travel the intervening distance? Actually, for a 2.0 Solar Mass star half of this mass (1.0 solar mass) of that has already been lost into interstellar space and is too far away to be illuminated by the central star when it is finally exposed!

    A general standard reference for this is Iben Jr’s;
    The life and times of an intermediate mass star – In isolation/in a close binary

    The relevant section is 4.1 Mass loss (pg.19.), especially;

    “From the standpoint of its influence over the final fate of an intermediate mass star, an overwhelmingly important feature of the AGB phase of evolution is mass loss from its surface.”

    Iben also says (pg.24);

    “When a star leaves the AGB as a consequence of superwind mass ejection, the mass of its remaining hydrogen-rich envelope is of the order of one-tenth of the mass….

    This infers that most of the mass loss occurs during the AGB phase. As Fig. 16 (pg.25) points out;

    “The evolutionary track might be appropriate for a star of initial mass about 2 Solar Masses which loses about 1.2 Solar Masses during the AGB phase via an ordinary wind and then ejects a nebulae shell of mass about 0.2 Solar Masses after developing a C-O core of mass about 0.6 Solar Masses.”

    The illumination of the Planetary Nebula occurs after the ejection of the shell when the exposed core can UV fluoresce the expanding shell. Hence, most of the ejection has occurred BEFORE the appearance of the planetary nebula.

    NOTE: As for quoting the Wikipedia article is quite extraordinary. I actually wrote this originally for this Planetary Nebula article, and I even modified this exact same sentence on 9th May 2010!

  17. “I acknowledge that my preference for 3 instead of III (etc) is atypical.”

    Yes, it is confusing to say the least. The use of I and II for these populations of stars was first adopted from Walter Baade’s paper “The Resolution of Messier 32, NGC 205, and the Central Region of the Andromeda Galaxy”; that appears in Contributions from the Mount Wilson Observatory and the Astrophysical Journal, Volume 100, 137-146 (1944).
    Originally the difference in these two groups was by the galactic velocities NOT the ‘metal’ content. The galactic velocities for them were decided by Oort in 1926, but Baade found this was better correlated with the spectral class.
    According to Gaposchkin, the roman numerals were used as to not mix up the designation with the decimal system of spectral classes. I.e. G1 or G2 stars. [This was also before the formal luminosity classes which are roman numerals; I.e. Classes I to VI.]
    Traditionally it is Pop I, II, and now III.

  18. @ Hon SBC

    I liked the wikipedia article.

    My statement ‘blowing off a planetary nebulae’ is clearly in the context of seeding the interstellar medium with heavy elements – which you note as a key role of planetary nebulae in paragraph 3 of your wikipedia article.

    I feel it’s unreasonable to suggest my article is ‘wrong’ for failing to include 500-1000 words of minor qualifications, when we are in apparent agreement on the key point here.

    If we follow your logic – should we also include stellar wind before the AGB phase which also represents a substantial mass loss during the star’s life.

  19. Steve asked me for references. Now I said;

    “The article says;
    “Theory has it that Population 3 stars formed from a homogeneous interstellar medium of hydrogen and helium. The homogeneity of this medium meant that any stars that formed were all massive – in the order of hundreds of solar masses.”
    This is also quite wrong.

    Why this is wrong is because it is believed that the first stars formed along filaments like pearls on a string. I would have to do some searching for original papers on this, but recent one appears by Madau, P., et.al. “Radiative Transfer in a Clumpy Universe. III. The Nature of Cosmological Ionizing Sources”, Ap.J., 514, 648 (1999)
    When the first galaxies formed they were probably in a foam-like structures (as seen in the comic background distribution by COBE), with numerous voids and walls. It is within these long extended boundaries that the first stars sprung forth. Hence the original gas was far from homogenous – but more like many “co-joined mega-nebulae.”

    So which came first, the Pop III stars or the galaxies?

    Further, when the Pop III sprung up, they did so prodigiously that the initial starburst made them highly luminous. After which, the gravitation of the stellar sources reorganised the gas into the galaxies we see today. The gas was then mixed with ‘pollution’ produced by the Pop III stars, to be then recycled into the next generation of stars. This cycle then keeps continued increasing the average metallicity of the Galaxy.
    However, Pop II stars might form in early in the Universe, but it depends where the stars are formed – meaning Pop II stars might still have formed in the first billion years or in the last few billion.
    Hence Pop II are a gradient of metallicities and not just some single arbitary figure. I.e Z>10^-5 to say Z<0.01.
    Also the proportion of old stars to new is dependent of the distribution of the hydrogen. Elliptical galaxies and globulars seemed to have used all of the gas in star generation in one burst, while spiral galaxies had enough gas remaining so the recycling process could continue up to today. This suggests the distribution of the original hydrogen was also far from homogeneous. [Many of these ideas first appeared after about 1986 with the work of Joseph Silk. These same ideas, though more refined, still appear in recent works.
    Hence my objection here.

  20. “I feel it’s unreasonable to suggest my article is ‘wrong’ for failing to include 500-1000 words of minor qualifications, when we are in apparent agreement on the key point here.”

    You’re probably right in saying this, and there is little doubt of the background required in stellar evolution subjects. As ‘stellar evolution’ is often thought as “not cool” it is important to be precise in what you say – especially as lots of kids are learning about it in school or college. Clearly reading the responses here (and in other UT articles with stellar evolution being discussed) shows the wide gaps in people’s knowledge. I’ve actually taught stellar evolution classes over the years – mostly aimed to plug these same inconsistencies in the available books, articles and the literature.

    So though whilst it is not probably that important to be exact, it is also important to paint an accurate the general picture.

    Note: Perhaps my objection is more with the completely wrong concept that planetary nebula are one solitary explosive catastrophic event – like all novae or supernova – where “blowing off a planetary nebula” just infers and reinforces that exact view. (This old view of planetary nebulae died in the 1960s!) The visible nebulosity is mostly because of the earlier AGB mass loss by winds from the stellar atmosphere, which for a solar-like one mass solar star continues for about 20 to 50 million years! Formation of the observed nebulosity is by florescence from the UV energies of the central PNN. It is short lived – akin to the last breath before the stellar core is exposed and the star dies.

    A better statement would have been;

    “And when I say death throes seeding the interstellar medium this includes average sized stars losing half of their mass blown off by strong stellar winds during their red giant phase, novae – or bigger stars exploding as supernovae.”

    This is my point here.

  21. Steve wrote;

    “My statement ‘blowing off a planetary nebulae’ is clearly in the context of seeding the interstellar medium with heavy elements – which you note as a key role of planetary nebulae in paragraph 3 of your wikipedia article.”

    I actually didn’t write this as said, as it is slightly modified by other wiki contributors. However, the Second paragraph clearly states what I’ve again just stated above.

    I will tighten the Wiki text so it is a bit clearer, stating; “Planetary Nebulae show us the crucial evidence that dying stars do play a role in the chemical evolution of the galaxy….” (It is a excellent example of imprecision in the wording.)

  22. I cant resist.
    So from a layman’s perspective. 🙂

    Population III (3) stars might not be stars at all by our current definition. More like Bright spots in the Clumps of Hydrogen Gas Remaining after the Big Bang that burnt out quickly then exploded.

    In exploding, they left holes in the clumpy gas clouds and energized them enough so that turbulence created Population II stars?

    I will evoke Occamz here as my justification and say that it sounds like a galaxy making process to me.

    But I’m an artist not a scientist.
    More right brain, less maths. 🙂

    As for the future of stars and the heat death of the universe, well thats a sad outlook. But does make sense.

    From a creative point of view I sometimes think of all this energy in the universe as a factory and I wonder what will remain once its all cooked and cooled down.

    Will gravity remain or will mass no longer be a measure of its energy equivalence? Will heat death mean that the universe burns up all its mass? heavy elements included. Will it just all be dark matter remaining. (pun intended)

    Will black holes remain in the cold universe.?

    So many unanswered questions. 🙂

    Damian

  23. Since my previous post is in the ever-lasting moderation-cycle (if it will ever leave that path, please delete this comment 😉 ), I post it again, but with only one URL:

    @ Damian

    May I guide your attention to this book:

    Death from the skies! by Phil Plait from badastronomy.com .

    Its last chapter covers all the questions you have, and the whole book is really fun to read.

  24. The previous discussion is pretty illuminating [sic!], thanks all.

    I will evoke Occamz here as my justification

    You could do that, but I note that Occam’s razor in a falsifiability theory of science isn’t useful for “justification” (which then is arbitration by testing). But for posing likely and robust theories in the first place, and for temporarily choosing between equally predictive theories in the unlikely event the state of the art comes down to that.

    The nitpick point being that if it is “justification” at all (doubtful IMO) it is pretty weak.

    More right brain, less maths.

    Another care/nitpick: The folk psychology general right-left dichotomy has IIRC been falsified. Brain processing is more complicated than that, shuttling information to and fro between centers.

    Wikipedia to the rescue: “Broad generalizations are often made in popular psychology about certain function (eg. logic, creativity) being lateralised, that is, located in the right or left side of the brain. These ideas need to be treated carefully because the popular lateralizations are often distributed across both sides.[1]”

    Sorry, that is what happens when commenting on science blogs; people will pick your brain … ehm, nits!

    From a creative point of view I sometimes think of all this energy in the universe as a factory

    Well, energy in itself doesn’t do anything. And in fact the universe eventually sums to zero energy flat space.

    It is the different energy distributions which do. Thus entropy, which tells us how energy distributes, is what runs the universe. (Runs it down, more precisely.)

    So basically it is the universe expansion that allows for entropy and a time arrow. Then inflation is the factory that both makes universes happen (multiverse theory) and makes them “happen” (essential in big bang). 😀

    Will black holes remain in the cold universe.?

    I believe the current consensus is, since Hawking conceded his bet that touched this, that they eventually evaporate when the universe disperse towards heat death (max entropy) as the last objects actually doing anything interesting.

  25. Although LBC didn’t quite get how the early universe formed very well… his points on the first stars actually isn’t far off.

    In a nutshell: after the big bang, the primordial soup was rather smooth (this has been proven); this soup was made up of H, He, dark matter/sub atomic particles (comprising 9/10th’s of the matter).
    How the Universe went from this smooth dark period, to the network you see in the early cosmic background picture isn’t known for sure… altho there are several theories. Reason is, we know the early dark period consisted of a smooth and even universe; however, if the soup was consistently smooth, gravity would have maintained this order indefinitely.
    The leading theory is, there were small inconsistent areas with very small variations in density. I think was LBC was trying to say on this point, is while particles of ordinary matter readily interact with each other… and if electrically charged react w/ EM radiation. Where as dark matter is comprised of particles which do not react with such radiation, yet DM does act gravitationally like any ordinary matter.

    So in theory, gravitational attraction caused dark matter density variations to condesne into a network of filaments and concentrated areas. In theory… unlike normal matter, DM cannot or mostly did not collapse into dense objects like stars, brown dwarfs, and stellar remnants. Interestingly, dark matter would have been at the center of star forming regions in the early universe, where as today it sits outside… on the edges of galaxies.

    As far as the first stars… there were likely some which did become black holes, there may have even been small black holes created by the big bang. There is no doubt the first stars were larger than our sun… exactly how much larger is a bit of debate. It all comes down to what temperature you believe the Universe was at during this period (IMHO there were variations). How we come to this in a bit…

  26. To continue… just 300 million years after the big bang…
    Early stars were made of hydrogen (duh moment), so the proof comes down to how this element reacts; in this case into the first star forming regions.
    How the universe cools plays a large role in how ordinary matter in the primordial system to separate from dark matter. Hydrogen (when cooling) would have setteled into a pancake shape, rotating configuration to form something clumpy and disk like. Because DM particles would not emit radiation or lose energy they stay scattered in the primordial cloud. This star forming region settles down to look like a tiny galaxy with a disk of ordinary matter and a halo of dark matter. Within the disk the densest areas of gas, clumps would continue to contract and eventually become stars.

    Now to size… The first star forming clumps were warmer than the molecular gas clouds today’s stars form in. Today these clouds consist of dust grains and molecules of heavier elements which act to cool the gas cloud. The smallest mass a clump of gas must have to collapse under its own gravity is known as the Jean mass.
    Jean mass is proportional to the square of the the gas temperature…. and inversely proportional to the square root of the gas pressure. Although the first stars would have had approximately the same pressure as today’s gas regions, the temperature of the first collapsing gas clumps were nearly 30 times higher; this comes down to having a Jean mass around 1,000 times higher.
    Working out the math with the Jean mass of today’s star forming molecular clouds in our own galaxy, and scale this up by a factor of 1000, we come out with an estimate of the first stars being 500-1000 times the mass of our Sun. Some computer simulations have come up with more massive Population III stars.

    Early black holes are another lecture, but I believe some of these early stars would have been in some very dense areas to collect enough mass inside its core to collapse in on itself, instead of blowing itself apart.
    It may even be possible some of these Pop III stars fed on surrounding gasses to become large enough to become supernovae.

    There is radical thoughts about dark stars forming just 100-200 million years after the big bang…which is sort of a fascinating idea. This would toss current stellar evolution on its head.

    When it comes down to it.. there is a lot of room for speculation. Most of today’s acceptance of theories is from computer simulations. Which are becoming better all the time… but they aren’t perfect. Hopefully with new technology like LOFAR, the new radio arrays being constructed, and even the LHC, we may finally have a lot more proof and less speculation.

  27. All non-spam posts have now cleared the moderation phase (whether they are Pop III, or 3, posts, I do not know).

  28. Aodhhan, What I discribed is a back of the envelope description of inflationary cosmology. What I wrote is a sketch of canonical theory.

    LC

  29. LBC… YSR.
    This is another ramble where you use the words, but not correctly; or in this instance, incomplete.
    So it appears your 100 level of this subjecct is equally shared when using the term canonical. So mixing them is bound to cause problems.

  30. This recent preprint may be of interest (link is to abstract): Population III Star Formation During and After the Reionization Epoch.

    Here’s the abstract:

    Population III star formation during the dark ages shifted from minihalos (~10^6 Msun) cooled via molecular hydrogen to more massive halos (~10^8 Msun) cooled via Ly-alpha as Lyman-Werner backgrounds progressively quenched molecular hydrogen cooling. Eventually, both modes of primordial star formation were suppressed by the chemical enrichment of the IGM. We present a comprehensive model for following the modes of Population III star formation that is based on a combination of analytical calculations and cosmological simulations. We characterize the properties of the transition from metal-free star formation to the first Population II clusters for an average region of the Universe and for the progenitors of the Milky Way. Finally, we highlight the possibility of observing the explosion of Population III stars within Ly-alpha cooled halos at redshift z~6 in future deep all sky surveys such as LSST.

  31. Thank you all for answering some of my questions.

    @ DrFlimmer

    I didn’t know Phil Plait wrote a book, awesome, will definitely look it up.

    @Torbjorn Larsson OM:

    Cheers, for that 🙂 Good to know I’m not left brain deficient, just math deficient. Apologies for invoking Occam’s Razor, seemed like an appropriate choice of words at the time. Let me rephrase. ” Its a concept that seems comprehensible” 🙂

    Wow, so the current consensus is that even black holes will succumb to the heat death of the universe. Thats Sobering.

    I think its important for all scientists to realize that the theories put forward are considered for their Philosophical context by some of us. These are big questions. Inquiring Minds will seek answers, and try to contextualize it on human scale. (much lesser ideas have spawned religions in human history)

    And the Internet is full of armchair Philosophers. 🙂

    Its hard to imagine the nothing that preceded the universe let alone the nothing that will be left.

    Damian

  32. @ damian

    Phil Plait actually wrote two books (the link above is to his second one). And the phrase “Phil wrote a book?” is something like a running gag on BadAstronomy. 😉

  33. “Although LBC didn’t quite get how the early universe formed very well…”

    I thought LBC’s post was clear, concise, grammatically correct and to the point. Also, crucially, he described the role of inflation in the early universe, something you failed to mention at all!

    “This is another ramble….”

    Compared to your two long, rambling posts? Especially this gem:

    “How the Universe went from this smooth dark period, to the network you see in the early cosmic background picture isn’t known for sure… altho there are several theories. Reason is, we know the early dark period consisted of a smooth and even universe” Huh? How?

    Your using non-standard terminology AND not making sense. You might want to tighten up your own posts before casting aspersions on others.

  34. Aodhhan said; “There is no doubt the first stars were larger than our sun… exactly how much larger is a bit of debate.”

    Utter rubbish. Where is the evidence for this?

    The current evidence is that the first stars ranged between the small to very larger stars. The difference is that stars behave differently because there are no ‘metals’ mixed in the available nebulosity.
    How do we know this? Mostly because Pop I and II are chemically different and evolve differently. I.e. The observed luminosity differences between the Pop I and II’s.
    Hence, if the Pop III stars near the beginning universe are more pure, we can extrapolate their general behaviour.
    Cosmologists are interested in the very large Pop III stars because observing them as supernova as they might be observable at the very high redshifts. The largest of them might also explain the massive black holes in galaxy centres.

    At such large cosmological distance it is unlikely we will ever gain evidence of then smaller Pop III stars, except perhaps for the white dwarf remnants.

    As for the rest of your diatribe, again you just aim to attack Lawrence in an attempt to make you feel more intelligent, and instead your mishmash of words really make you look like foolish.
    Frankly, you are clearly well out of your depth (yet again.)

  35. Remember that the reason for why pop-III stars are suspected to be able to become more massive is the fact that the main energy source, hydrogen burning, will be forced to use the less effective p-p chain, instead of the faster (at high temperatures and densities) C-N-cycle. The less effective burning makes the Eddington-limit larger (more mass), allowing for a more massive star to be stable atleast during main sequence. Even small pollutions of carbon or nitrogen into a star makes alot of difference here, causing analogies and extrapolation from now common stars to be difficult, even dangerous.

  36. I am writing a bit out of my field when it comes to popIII stars, but here goes with what I know. There is some kurtosis in the statistics in the CMB anisotropy which correlates with some very early stars. As yet these are too far and weak in the IR to image. Near IR and optical work is needed to push this envelope.

    There is some debate over whether these stars are highly massive or not. So I can’t say with much authority on this. However, here is my sense of things. As Excalibur points out these stars were likely powered largely by p-p fusion, but this competed with p-D fusion. Recall that deuterium was generated not in stars but in a fairly early phase of the universe where protons and neutrons went from being in a state of equilibrium through collision to out of equilibrium. The computations on this were worked out in the mid 1970s, and theory is supported by observation. So a part of the problem is the energy mechanism. The problem is that a near zero metallicity star will have far less opacity. So the star may require a large amount of mass so the small photon scatter can provide some pressure against gravitation. These stars may not have ever been particular stable, forming as the implosion of matter anisotropy and leading to a runaway fusion process which exploded. A low mass popIII star seems to require some sort of stability, even if for a few 10’s of millions of years. To be honest I suspect this is hard to model.

    LC

  37. PS, I forgot. I think there is evidence of considerable mass as well. These stars produced a copious amout of X-rays that ionized matter. The universe was largely a dilute gas of ionized hydrogen, and this is still a prevalent condition. That took a hell of a lot of X-ray energy!

    LC

  38. @ damian:

    Um, yes; but I believe DrFlimmer were more helpful to you. That book* will likely stimulate all of your brain, if the rumors are correct.

    [* “Phil wrote a book?” ]

    Its hard to imagine the nothing that preceded the universe let alone the nothing that will be left.

    On the nothing that preceded the universe:

    Here are two physicists working it out for us.

    Sean Carroll finds it a meaningless question; “nothing” isn’t a member of the observable distribution of “something” (distribution of universes).

    Vic Stenger notes that spontaneous symmetry breaking is observably what will happen in systems, and that tending towards nothing means having more symmetries. So if there was a “nothing” it was unstable.

    If you are concerned that his particular pity example, that is linked to in the post, starts out with laws, he is more general in his book “God – the failed hypothesis”. Essentially laws are the remaining symmetries and broken symmetries that remains after the above process. (Say, conservation of electromagnetic charge, a law, is due to a symmetry of the electromagnetic field.)

    [My personal view is that Stenger’s hypothesis is inclusive and falsifiable, while Carroll’s is exclusive and his exclusion can never be tested.

    But IMHO Stenger’s hypothesis is not enough, since it is possible and even likely that inflationary multiverses are effectively eternal. It would help make such a theory consistent on _possible_ pathways, but it wouldn’t treat actual _likely_ pathways taken. Long story, I’m afraid…]

    On the nothing that will be left:

    Well, if inflationary multiverses are correct, YMMV, the multiverse will continue as before. Younger universes will be where the action is.

    Our universe to younger bubble universes: “-Hey! Get off my lawn!”

  39. @LBC

    What runaway fusion process? Im not sure at what mass-range you are referring to, but small stars will be supported by normal gas pressure, above 20 or so Msun radiation pressure are becoming increasingly important. But small stars will have a very hard time to form in the inital stage, they might not even be able to form before being afflicted by the effects of a large mass star having its way with the environment. The only runaway fusion process i can imagine would be the pair-instability effect that is suspected to be enabled above approximately 100Msun.

    Jeans-mass is very an important factor here. With a cooling medium, the Jeans mass becomes smaller and smaller and eventually reaches a point where there actually is some areas where the gas cloud can collapse on itself. At the same time the Jeans mass assures that below that scale the gasdistribution is mostly smooth because it cannot collapse on itself. So on galactic scales cosmology tells us about a filamentary structure, whereas on a stellar scale the distribution is assured to be smooth, and the scale is slowly moved down by cooling. Eventually this scale becomes on the same order as the Jeans-mass and stars will start to form, naturally at the high end

    The forming protostar will burn traceamounts of Lithium and Deuterium in a rather rapid process, but before the helium burning is initiated those reserves will have been used up. The lower convection expected in such stars will allow for shell-burning of lithium.

    As far as i can see (and that may be limited) the end result would be stars with a massrange (that i dont know exact, but both lower and higher end cutoffs) that is somewhat more dense than normal stars today, hence it will also be hotter on the surface than a simliar star today. The upper end of stable stars as dictated by the Eddington limit is significantly higher, until the local volume have been polluted by metals.

  40. So if there was a “nothing” it was unstable.

    Actually, it was maximally unstable. It is fun to ponder if “nothing” was so unstable, so the likelihood of it ever existing was efficiently “nothing”. Sort of “a probability singularity”. 😮

    [This is possibly more fun in swedish, since one of our early television age comedians wrote a famous if dated monologue on nuclear safety, pulling statisticians’ legs; roughly:

    “Likelihood, what, it means something that is like truth. But really as like truth as truth it isn’t, if it is likely.

    Now, we can’t afford genuine truths anymore, but have to make do with probability calculations. Pity, because they have lower quality than truths. They aren’t as reliable. For example, they become very different before and after.

    I mean before Harrisburg it was very unlikely that what happened in Harrisburg would happen, but as soon as it happened the likelihood shot up to become no less than 100 % so it was nearly true that it had happened.

    But only nearly true. That is what is so peculiar. It is as if they mean that what happened in Harrisburg was so utterly unlikely so it really didn’t happen. …”]

  41. Excalibur,

    This area is a bit out of my domain. Yet PopIII stars don’t obey the standard rules quite in the same way as PopI-II stars do. The Eddington limit equates the gradient of pressure with static g*d, d = density, and GMd/r^2 for gravity. The differential pressure depends linearly on the opacity as dP/dr = k*d*F, where for pure hydrogen this small. The radiation force F is written according to the luminosity so that L ~ 1/k. So the small opacity means that the luminosity is enormous and stability is questionable. I can’t comment on Jeans mass in this content, for it involves some compuatations of number densities, and right now I don’t know how that applies for pure H with this funny balance in hydrostatic pressure with low opacity.

    This might mean that a popIII star accretes mass almost endlessly and hydrogen falls on it and the luminosity might grow to enormous values. This is a different sort of behavior from popI-II stars. These stars reionized the entire universe, so it seems to me these generated far more energy than any regular popI-II star (huge power output) in order to do that. Again I am a bit out of my element here, and I am not aware that any “standard model” for popIII stars has been agreed upon, and certainly it seems to me these stars are not well understood systems.

    Torbjorn Larsson OM , The observable universe may not have emerged from absolute nothing. The vacuum energy which formed the basis for inflationary cosmology in the first 10^{-20} seconds of the universe may have tunneled from the vacuum of another cosmology. Since this inflationary vacuum was so enormous, 10^{123} times what the vacuum energy density is now, this probably tunneled from another cosmology near the singularity of a black hole there. So I offer up a plausible scenario for this, which might (if I am sufficiently on the mark — which is at best a maybe) be some approximation for this.

    Quantum tunneling usually involves how a quantum particle may traverse a tunneling region with a potential energy larger than the kinetic energy of the particle. The Schrodinger equation (SE)

    Ihbar &Y/&t = (?^2/2m)&^Y/&x^2 + V(x)Y

    (Y = “psi” the wave function) in one dimension is the classical example in Mertzbacher and other texts. For a stationary phase Y(x,t) = Y(x)exp(-iEt/h-bar) and we have a basic position dependence with Y(x) ~ exp(ikx), the SE is easily seen as

    EY = (hbar k)^2/2mY + V(x)Y,

    And the solution for the wave vector k or momentum p = hbar*k is

    p = sqrt{2m}sqrt(E – V(x)).

    The tunneling probability may be explicitly computed by knowing the form of V(x) and working out boundary conditions, which is not conceptually difficult but a bit tedious to work through. The form of the momentum p here though is imaginary if V(x) is larger than E, and the tunneling probability is greater than zero. Now the form of the wave function with this imaginary p is of the form Y ~ exp(-|p|x/hbar), which for the magnitude |p| very large (equivalently large V(x)) is a rapidly dropping to zero exponential. So we don’t expect a significant tunneling process.

    If the potential is very large at its peak V_{max} ~ 2mc^2 for m the mass of an electron there is a probability that the e-e^+ pair created here will annihilate the e^- at one side of the potential with the e^+ and the pair generated e^- escapes to the other side. Since this is a quantum process then what ever information is carried by the initial e^- is the same as on the e^- which has tunneled through. This is a sort of resonance phenomenon.

    So what does this have to do with cosmologies and the landscape? Hawking, Halliwell and others proposed a version of the “no-boundary cosmology” where a spacetime cosmology with a particular arrow of time and a CPT violation (say left handed) is mirrored by another cosmology with an oppositely directed arrow of time and a CPT violation which is opposite (say right handed). So we might think of these as a cosmology and “anti-cosmology” with opposite quantum numbers or topological indices and …, all which make things cancel out to zero. So we then have a huge landscape with a large potential barrier. The basic Friedmann-Lemaitre-Robertson-Walker metric has a tunneling barrier, which is a local aspect of the more general landscape potential barriers. So the symmetrical portion of a cosmology can be then this curious anti-cosmology.

    Now consider a region near a black hole singularity. There the tidal forces of gravity are enormous and a wave function is then squeezed. The phase space volume it occupies is “squashed,” and this means its uncertainty in certain directions becomes very large while the conjugate momenta uncertainties becomes very small. As such a patch or region near the singularity is sufficiently quantum uncertain that it may become lost in this quantum noise. In effect this patch and the quantum vacuum energy it contains has quantum tunneled out of the universe which contains this black hole, or it “ventures” into this potential hill. For a large potential hill there are virtual quantum pairs of these cosmologies (universe plus anti-universe) or a virtual biverse. Will this little patch or bubble of vacuum energy near the singularity has some probability of annihilating with the anti-universe which then lets the virtual universe escape beyond the potential well. This patch or virtual bubble of spacetime then becomes the “seed” for a nascent cosmology.

    Now connect up with my description above for inflationary cosmology, where this tunneled vacuum inflates cosmology and generates small anisotropies and inhomogeneous distributions of matter. Then segue into the problem of popIII stars and then … . A lot of work is required to connect some dots here!

    LC

  42. @LBC

    Thanks for the information, interesting insights. I am not sure i follow how radiation pressure at low opaquity would possibly lead to almost endlessly massive stars. I understand it will allow for larger stars, just that i dont see it lead where you say. But i will look into it deeper to understand it in more detail.

    Thankfully stars will never be made of 100% hydrogen, thanks to comsology they will always have atleast ~25% helium and some trace amounts of lithium, beryllium & borium. Not sure if that changes anything, i suspect this is already what you meant…

  43. @Excalibur,

    I am not certain about this, but I have a physical intuition about this. The low opacity means that lots of X-rays escape these stars before contributing to the hydrostatic balance. In fact this article appears to suggests that e-e^+ production plays a role in this process. So a smaller clump of hydrogen might compress a whole lot to be stable, while a larger mass might not have to. The larger popIII star has more mass to capture X-rays and “use them” for hydrostatic equilibrium.

    This is just my physical sense of the problem. I am sure that people are running sophisticated models on supercomputers to look into this.

    LC

  44. PS,

    If I am right this sort of puts the main sequence idea of stars on its head. Smaller mass popIII stars might actually be hotter (I mean really hot) than more massive ones. I am not an expert here, but I think with popIII stars you are out of the Kansas of main sequence logic.

    LC

  45. Lawrence B. Crowell, universe archeology however interesting is off topic on a thread of stellar archeology, so I will be fairly brief. Yes, since open FRW universes are zero energy by their dynamical system behavior, they may tunnel from earlier preuniverses:

    “Indeed, since the universe is by definition an isolated system, the zeroenergy result is compatible with the universe emerging from a ‘‘ system ’’ with zero energy, be it quantum vacuum (Albrow 1973; Tryon 1973; Guth 1981), ‘‘ nothing ’’ (Vilenkin 1983), flat empty space (Prigogine et al. 1988, 1989; Gunzig et al. 2001; Saa et al. 2001), or something else.”

    [These authors also note that the reverse, a universe out of a non-zero energy system seems thermodynamically inconceivable.]

    However, in my comments to damian I tried to reply to “the nothing that preceded the universe”.

    This is the “something out of nothing” question that bothers some. It typically will have people claim that answering with an already lawful preuniverse is either “a category error” (if they are philosophically minded) or, more appropriately, “merely pushing back the problem on an earlier universe” (if they are empirically minded). So it merits an explicit answer.

    [FWIW, I happen to agree with the later group in that preuniverses, be they the inflationary multiverses that the standard cosmology may lead up to or “no boundary condition” preuniverses, are inconsistent by way of such incompleteness. They don’t predict all what they need to predict to predict actual universes.

    This is where Linde’s inflationary cosmology, with some patching of pathways to satisfy necessary anthropic likelihood criteria, satisfies. (Me, at least.) As I said before, it is a long story, albeit I believe I’ve described most of it on UT in earlier threads.]

  46. Btw, some of you touched structure formation above.

    FWIW, today I stumbled on this arxiv review of current galaxy formation theory, which briefly sets that scene (ch 2.2, p3). A hierarchy of models and empirical methods gives results (“a cosmic web”) which “is consistent with measurements of galaxy and quasar clustering on a wide range of scales.”

    [The paper ends up arguing for such an hierarchal structuring of galaxy formation research; out of necessity.]

  47. Torbjorn Larsson OM: The k = 0 FLRW cosmology has in some definition a zero energy. Mind you, energy is a tricky concept to work with or define in general relativity. In fact for cosmologies there are no time-like Killing vectors which define an isometry for the conservation of energy. Further, prior to the universe assuming am FLRW configuration it was a de Sitter vacuum, and the transition to reheating at the post-inflationary period changed this dynamics. Curiously, with the lower vacuum energy (dark energy) the universe began to transition back to a de Sitter configuration about 4 billion years ago.

    I did mention this, one of course because it is my interest, but also that these popIII stars fit in a chain of astrophysical or cosmological history, or causality, that is unknown. This is the so called dark age of the universe, which ended with the reionization of luminous matter. This looks to be the last optical or IR frontier for optical astronomy with regards to cosmology.

    LC

  48. @LC

    Can you please elaborate on what you consider being small stars here and what you consider large?

    The reason i ask is that if it is really small stars, then they reach hydrostatic balance by normal gas pressure, wich makes opaquity less important (it still is important in terms of the temperature ofc). And as the article mentions, pop-III stars are expected to all be above 100Msun.

    I agree that a 100Msun+ star made of only pristine cosmological soup will never entirely fit into something like the main-sequence.

  49. Oops. looking back, some awful errors here:

    – Universe “archeology” _may_ be OT. Tough call.

    – “They don’t predict all what they need to predict to predict actual universes.” Um, no, they do that all right. I’m not sure how to put this at the time. Bother; it is usually me arguing that eager predictiveness is all that matters. :-~

  50. Lawrence B. Crowell, yes on the GR/de Sitter et cetera. But I’m not sure why you raise that since the paper explicitly treats all of that and shows how you get around that by treating it as a dynamical system. The GR problem goes away as it is internal to GR and the de Sitter problem goes away since it is explicitly treated.

    The point is that _all_ open FRW universes are zero energy in a proper sense to understand universes behavior (including flatness), if the paper is correct.

  51. This definition for energy comes from the T^{00} part of the stress-energy tensor. These are components that are determined in a frame. To actually calculate energy you have to evaluate the tensor in a spatial volume by a Stoke’s law calculation. For various reasons this is a subtle issue. One is relying upon the continuity of the stress-energy tensor, nabla_aT^{ab} = 0, but the basis vector e_b that one contracts with the tensor only gives a reasonable calculation if it is a Killing vector. Cosmologies of this type have no Killing vectors. This leads to a very interesting question or set of issues as to what we mean by energy in a cosmology.

    LC

  52. @Excalibur, Again I can’t write with much definative authority. I also have not sunk my teeth into this problem. The “equation of state,” if you want to call it that for popIII stars, is different as I see it. The low opacity suggests one needs to have very high densities and pressures in these stars. So for a lower mass popIII star I might think that they have to be very compressed to remain at all quasi-stable. This would mean they are extremely hot. A larger mass popIII star would have larger luminosity but has enough hydrogen to permit photons to exert some outward pressure without exceedingly high temperatures.

    This is a physical argument, and as I said I have not dug into this in detail. I could very well be wrong, and those enterprising enough might want to try to demonstrate that I am wrong. These are very odd stars in my opinion.

    LC

  53. @LC

    Not trying to prove anyone wrong, just trying to find more solid information, and that does include questioning information that isnt immediately obvious.

    Pop-III stars are indeed odd beasts, they have to be, i already realised that long ago, but still they are not magical beings, they can be understood in terms of physics.

  54. I have not been able to reply to the Pop III article, but I am still surprised at the responses here.

    The problem of Pop III stars has been investigated since the 1970s, whose stellar evolution became seriously questioned and generally solved in the early 1980s. The concept of Pop III stars were first considered very similar to what we know today, whose limitations have been mostly observational evidence. There existence were not first considered by cosmologists but only by those doing stellar evolution work. It was also realised the role of metals / Z made these stars evolve differently, and were considered to be mostly small to very massive stars. It seems that the low-metallicity problem with purely hydrogen and helium (H, He) is the larger size of the core compared to stars seen today stars of today.

    For small stars between 0.8 and 5 solar masses, the theoretical conversion of hydrogen in the H,He mixture, is that these stars probably only make Helium and produce very time amounts of metals. It appears when you up the size of the star above 20 to 500 solar masses that any Z metals are created in sufficient quantities to pollute the universe for the coming many generation of stars. The idea of this began in about 1977. Most of the massive stars are considered all to explode as supernova, or after about 1979, the so-called PCSNe – pair created supernovae. It is believed the outcomes of these stars were strongly enriched with carbon and oxygen (with mostly carbon-oxygen cores.) There is also little doubt that these large stars are not one but several generations before the higher metallicity were high enough to make Population II stars. The best dual evolution papers from the 1980s that discusses this is probably Ober, W.W., et.al “Evolution of massive pregalactic stars I “, A.&A., 119, 54-60 (1983) and Ober, W.W., et.al “Evolution of massive pregalactic stars II. Nucleosynthesis in pair creation supernovae and pregalactic enrichment.”, A.&A., 119, 61-68 (1983)

    How these stars were formed was discussed as early as 1972, though much of this early work have be supplanted by more up to date studies, outstripping much of the paper and research that is being conducted today.

    Of course the interest today is the possibility of find PCSNs from large telescopes at the earliest times of the universe, probably because the should be far more frequent and outstandingly bright with +100 solar mass exploding stars. It is this reason why small Pop III stars are not of interest, because the 0.8 to 5 solar mass stars after 13 and a bit billion years are all burnt out white dwarfs. As these stars above around 1.5 solar masses are not very luminous, the only way we can detect them is if they are close to the sun. Else, we have looked for them in globular star cluster, and found white dwarfs, but their exclusive faintness means we cannot find spectrally if they show the pure signatures of Pop III white dwarfs. One possibility is we will see an ancient Pop III binary star which merges and produces a supernova Type I explosion – whose chemical signature would be different than a similar scenario for less ancient stars. (I read little about this scenario, but it does appear in the literature.

    Until we find such smaller Pop III or large PCSN objects, they will remain theoretical, but there is little doubt the they really exist. As we see stars in the universe, and we see stars in galaxies back to a couple of hundreds of million years after the Big Bang, the first stars have existed in this small gap.

    I have also given another two other interesting technical definitive papers in the follow posts to this one (we cannot it seems post two links in the one article in Universe Today for so reason or others – without being told to “Awaiting Moderation”. (This should fill in the gaps questioned by other bloggers in this post.

    Note: In my original post I spoke of the fact that the first stars were Pop III and larger objects. In the mid-1980s these massive objects were called VMOs (Very Massive Objects), though I do not think the name was formally adopted.

  55. Kashlinsky, A.; Rees, M. J., “Formation of population III stars and pregalactic evolution”, MNRAS., 205, 955 (1983)

    This is the definitive paper on the formation of Pop III and of the general nature of the massive objects. See especially on pg. 965, where Equation (25) sets a limit on the size of the smallest Pop III due to the way the disk theoretically collapses. If I can recall, this concept was ultimately rejected as the size of the protostellar fragments (actual smallest final Jeans mass of the nebulosity) and could behave quite differently than when they were first speculated.

  56. After reading the papers i find that http://articles.adsabs.harvard.edu/cgi-bin/nph-iarticle_query?1983A%26A…119…54E&data_type=PDF_HIGH&whole_paper=YES&type=PRINTER&filetype=.pdf hints that pop-III stars below about 60Msun probably do not form at all in the inital stage due to lack of cooling (metals). The upper limit from the papers seems uncertain, somewhere around 500-810Msun.

    Stars between 100-300Msun are expected to explode as pair-instability supernovae, stars below about 100Msun are expected to pulsate during the crucial moments and eject matter but survive for the time being and eventually collapse at a later stage, and stars above 300Msun are expected to collapse to black holes without a supernova ejecta.

    Since several generations of pop-III stars are expected before the first pop-II stars form, i suspect in a later stage there might be a pop-IIIa kind of lower mass star, but thats me speculating. Those would be mostly undetectable anyway at this time.

    This makes me believe that the 300+Msun pop-III stars would become excellent galactic seeds for SMBH, whereas the 100-300Msun pop-III stars would be the ones mostly enriching the environment with metals. The <100Msun stars would also enrich the environment, but with more carbon/oxygen and less heavier metals, end less total contribution (unless they are a much more common population)

Comments are closed.