Every now and then there are reports of a star that is older than the universe, or one that shouldn’t exist because it exhibits some strange property. It’s bad enough when the popular press makes these kinds of headlines to get pageviews, but it’s worse when the press releases themselves have such a title. That’s because most pop-journalism sites will simply copy-pasta the press release with little or no vetting, so a misleading press release spreads like wildfire.

Caffau’s star is a perfect example of this. Research on it was first presented in 2011 as an “impossible star” by ESO, and the story keeps returning with the a similar headline every year or so. What’s unfortunate about such headlines is that they miss the chance to look at what happens when observational data and presumed models come into conflict.

Caffau’s star is a star of about 0.8 solar masses with an extremely low metallicity. In astronomy, metallicity is a measure of how much “metal” a star contains, where metal in this case means anything other than hydrogen and helium. The ratio of metal to hydrogen is known as its metallicity. Stars can be categorized by their metallicity as population I (relatively high metallicity, like our Sun), population II (some metallicity) and population III (essentially none). While we’ve observed population I and II stars, we’ve never found a true population III star.

Such a star would be among the very first generation of stars to form. That’s because the early universe consisted of hydrogen, helium and some traces of lithium. Heavier metals were formed in the cores of stars, thus higher metallicity stars are generally younger than low metallicity ones. Caffau’s star has such a low metallicity that it is likely a second generation star, having formed soon after the very first stars experienced their explosive deaths as supernovae.

This makes the star unusual, but not impossible. The “impossible” part comes from the fact that it contains less lithium in its spectrum than we would expect for an early second generation star. While stars do consume lithium to make other elements (a process known as lithium burning) that isn’t enough to account for its lack of lithium. We’ve found other stars with similarly low lithium levels, which has given rise to what some call the lithium problem. Why do early stars contain less lithium than expected?

It’s currently an unsolved problem in astrophysics. Our estimation of primordial lithium levels from the big bang seem to be correct, and we aren’t sure how stars like Caffau’s star can have so little lithium. But that doesn’t make the star impossible, since it clearly exists. What it means is that our model is clearly incomplete, and that’s far more interesting.

The post And Yet, Here We Are appeared first on One Universe at a Time.

Much of the modeling of astrophysical systems focuses on dynamical behavior. That is, how stars, planets and interstellar clouds move under the forces of gravity. While such dynamical modelling can prove useful for studying the motion of galaxies, they are not quite as good at modeling the evolution of galaxies. That’s because galaxies evolve over time not only due to the motion of stars within them, but they also evolve chemically as old stars die and new stars form. The chemical makeup of new stars depends where and when they form.

One way you can look at the chemical makeup of a galaxy is to examine the metallicity of stars. The metallicity of a star is a measure of how much a star has of elements other than hydrogen and helium. Usually this is done as a ratio of iron to hydrogen, or oxygen to iron, for example.

In our own galaxy, stars with high metallicity tend to be near the central region and galactic plane, while lower metallicity stars tend to lie further from the galactic plane. This variation in metallicity is due to the fact that more stars lie near the galactic center and plane, therefore more stars die in that region, which frees up more metal (elements other than hydrogen and helium) to be used in new stars.

While it is fairly straightforward to model the broad trends of the chemical composition of stars with a galaxy, it would be nice to have better models. Enter a new attempt at a model integrating the dynamical motion of stars with their chemical evolution. The first results were presented in Astronomy and Astrophysics (paywalled, but the arxiv is here). In this paper the authors look at modelling the Milky Way.

To model the chemical evolution of the Milky Way, the team had to account for the fact that stars have the metallicity of their surroundings at the time and location of their formation, but also that stars tend to drift outward over time. So metallicity is a product of both star formation and stellar drift.

You can see the results in the figure here (from the paper). In the top row shows the stellar formation rate over time and you can see that star formation peaks at about 3 billion years and gradually tapers off. In the middle row shows a more detailed history, plotting the stellar formation rate over time at different distances from galactic center. The bottom row shows the metallicity in terms of the iron-hydrogen ratio and oxygen-iron ratio over at different times as a function of distance from the center.

The results are a reasonable match to the Milky Way, and this allows the authors to do a bit of detective work regarding the real Milky Way and our Sun in particular. Given the metallicity of our Sun, and its current location about 26,000 light years from galactic center, the authors calculated that our Sun likely formed at a location 18,000 light years from the center, and then gradually drifted out to its present location.

The post Chemodynamics appeared first on One Universe at a Time.

In the early moments of the universe, hydrogen and helium were formed through a process known as baryogenesis. Trace amounts of other elements such as lithium were also produced, but none of the heavier elements. This means that the first generation of stars were composed of hydrogen and helium, and it is only through fusion in their cores that the heavier elements we see today were created. The carbon, oxygen and iron in our bodies was produced through that process, which is why it is often said that we are star stuff.

Because heavier elements are only produced in stellar interiors, one way we can determine the age of a star is by the amount of these elements in its atmosphere, what is known as its metallicity. Younger stars will tend to have a higher metallicity, since they tend to form from the dust of earlier stars. Because of this, there has been a great deal of effort to discover very low metallicity stars. We have found some stars with quite low metallicity, indicating that they are second generation stars.

There has also been a search for first generation stars, but this has not been successful. Many of these stars would likely have been quite large, and therefore would burn through its hydrogen rather quickly, dying in a tremendous supernova explosion. One of the unanswered questions about these first generation stars is just how large they could get. The larger a star is, the shorter its lifetime. The general consensus is that first generation stars could have been as large as 200 – 300 solar masses, but proving that is a challenge.

Now a new paper in Science has found evidence to support this idea. The team looked at a low metallicity star known as SDSS J0018-0939. It is a dim orange star that seems to be a second-generation star. They looked at the abundances of 18 different elements within the star’s spectrum, and found that elements with an even atomic number (such as titanium and nickel) are more abundant than elements with odd atomic numbers (such as scandium and cobalt). This suggests that its progenitor star had some dying mechanism to produce more even elements than odd.

But this is something you would expect to see in the death of supermassive stars. When very large first-generation stars had fused most of their hydrogen into helium in their cores, they would then collapse quickly under their own weight. This would trigger a very large supernova (hypernova) known as a pair-instability supernova. In such a supernova, the gamma rays produced by nuclear collision are so energetic that they decay into electrons and positrons. These interact less strongly with helium, allowing more helium nuclei to fuse to produce heavier elements.  Since helium has an atomic number of 2, more even elements would be produced.

Simply finding higher even/odd element ratios in a single star isn’t enough to prove that first generation stars were quite large, but it does show that the idea is feasible. We’ll have to look at more second-generation stars to see if the idea truly holds up.

Paper: W. Aoki, et al. A chemical signature of first-generation very massive stars. Science, 345 (6199): 912-915 (2014).

The post Even Odds appeared first on One Universe at a Time.

In astronomy, all elements other than hydrogen and helium are referred to as “metals.” For this reason, a measure of the amount of other elements a star contains is known as its metallicity. One way to define the metallicity of a star is simply as the fraction of a star’s mass which is not hydrogen or helium. For the Sun, this number is Z = 0.02, which means that about 2% of the Sun’s mass is “metal”. Another way to express the metallicity of a star is by its ratio of Iron to Hydrogen, known as [Fe/H]. This is given on a logarithmic scale relative to the ratio of our Sun. So the [Fe/H] of our Sun is zero. Stars with lower metallicity will have negative [Fe/H] values, and ones with higher metallicity have positive values.When it comes to planetary systems, it’s generally been thought that planets would tend to form around stars with a higher metallicity. At a broad level that makes sense because rocky planets such as Earth can only form in a system where there are enough metals like iron, silicon, carbon and the like.  You can’t make a terrestrial planet out of just hydrogen and helium.  But now that we’ve discovered lots of exoplanetary systems, we can actually put this idea to the test. A recent paper in Nature has done just that, and they’ve found something rather interesting.

Exoplanets group into three types based upon stellar metallicity. Credit: Buchhave, et al.

In the paper, the authors looked at about 400 stars with exoplanets (about 600 exoplanets in all).  They then compared the size of the exoplanets with the metallicity of their stars.  What they found was that there was a distinct relation between the metallicity of a star and the type of planets it has.  Stars with a metallicity similar to our Sun’s were more likely to have terrestrial planets, while stars with higher metallicity tend to have gas dwarfs, or gas giants (Jupiter-like).  The study also showed that high metallicity stars are likely to have so-called “hot-Jupiters”. That is, large gas giants orbiting close to a star.

We’ve seen in computer models how large protoplanets will tend to migrate inward toward the star as they form. This new work would seem to support that idea, since higher metallicity stars would be more likely to form gas giants early on, thus allowing them to migrate inward to become hot Jupiters.

So it seems that metallicity is a significant factor in planetary formation, and higher metallicity stars will tend to form larger planets.  But it also seems that stars similar to the Sun are better suited for having terrestrial planets like ours.

Paper: Lars A. Buchhave, et al. Three regimes of extrasolar planet radius inferred from host star metallicities. Nature 509, 593–595 (2014)

The post Testing Metal appeared first on One Universe at a Time.

Last year a paper in the Astrophysical Journal announced the discovery of two super-Earths orbiting a low metallicity Sun-like star known as HD41248.  This was kind of a big deal, because it demonstrated that low metallicity stars could have rocky planets.  Most of the known exoplanets are around higher metallicity stars.  The two planets were discovered by observing the radial motion of the star (it motion toward or away from us) as measured by the Doppler shift of the starlight. Now a new paper in Astronomy and Astrophysics demonstrates that the planets likely don’t exist.

Radial velocity measurement for HD41248.
Credit: Jenkins, et al.

So how is it that a planet can be discovered one year and lost the next? It all has to do with the complexities of the observations. Exoplanets are not discovered by actually observing them directly. Of the hundreds of known exoplanets, only a few large ones have been directly imaged.  The rest have been discovered by their effect on the star they orbit, either by passing in front of the star causing it to dim slightly (transit method) or by measuring the wobble of the star as it is gravitationally pulled by the planet (radial velocity measure).

Of these two main methods, the transit method tends to be more reliable, but it only works if the planet passes in front of the star from our vantage point.  The radial velocity method can work regardless of the orientation of a planet’s orbit, but it has much more noise in the data.  That’s because the wobble of a star due to a planet is quite small, and with all the noise in the data you can sometimes get a false positive.

To measure the radial motion of the star, we measure the Doppler shift of the light emitted from the star. This light comes from the photosphere of the star. As the star wobbles toward and away from us, so does the photosphere of course, so by measuring this Doppler shift we can measure the motion of the star. But photosphere can have motion independent of the motion of the star.  Stellar activity such as starspots and flares can cause the photosphere to swell or shrink slightly, and that produces a Doppler shift as well.  Distinguishing between these different motions is a difficult challenge.

The original paper found signals of two planets, one with a period of 25 days, and the other with a period of 18 days.  Of these two signals, the 25 day period was the strongest.  The result was based upon 62 radial velocity measurements from the HARPS spectragraph, which is public data.  Since then, more observations were made.  The new paper looked at 162 new measurements, and from these determined that the 25 day period correlates with the rotational period of the star.  The 18 day period wasn’t clear in the new data.  So it seems that what looked like planets is instead stellar activity.

Just to be clear, the false positives of the first paper shouldn’t be viewed as a failure.  The results presented were well analyzed given the data at hand.  What this new paper shows is that the stellar activity of HD41248 is more complex than originally supposed.  This activity, combined with its differential rotation results in Doppler motion data that looks very similar to planetary influence.  This is cutting edge work, and as we do it we’re learning about where we need to be careful in our analysis.

Sometimes science is about making mistakes so we can learn from them.

Paper: J. S. Jenkins et al. Two Super-Earths Orbiting the Solar Analog HD 41248 on the Edge of a 7:5 Mean Motion Resonance. ApJ 771 41. (2013) doi:10.1088/0004-637X/771/1/41

Paper: N.C. Santos, et al. The HARPS search for southern extra-solar planets XXXV. The interesting case of HD41248: stellar activity, no planets? arXiv:1404.6135 [astro-ph.EP] (2014)

The post False Positive appeared first on One Universe at a Time.

After the big bang, the only elements in the universe were hydrogen, helium and trace amounts of lithium.  There was no carbon, oxygen or iron, because these elements are only formed when stars undergo fusion in their cores.  About 100 to 300 million years after the big bang, the first stars began to appear.  These first generation stars were likely very large, about a hundred times the mass of our Sun.  Because of their size, they had short lives which ended as supernovae.  From the remnants of those stars, a new generation of stars would form.  These second generation stars would have traces of elements such as carbon, but still lack heavier elements such as iron.  Now a new paper in Nature (arxiv version) has announced the discovery of just such a second generation star.

In astronomy, all elements other than hydrogen and helium are referred to as “metals.”  For this reason, a measure of the amount of other elements a star contains is known as its metallicity. One way to define the metallicity of a star is simply as the fraction of a star’s mass which is not hydrogen or helium.  For the Sun, this number is Z = 0.02, which means that about 2% of the Sun’s mass is “metal”.  Another way to express the metallicity of a star is by its ratio of Iron to Hydrogen, known as [Fe/H].  This is given on a logarithmic scale relative to the ratio of our Sun.  So the [Fe/H] of our Sun is zero.  Stars with lower metallicity will have negative [Fe/H] values, and ones with higher metallicity have positive values.

Spectrum of a low metallicity (Fe/H = -0.8) star. Credit: Anna Frebel.

Stars are often categorized by their metallicity.  For example, Population I stars have an [Fe/H] of at least -1, meaning they have 10% of the Sun’s iron ratio or more.  Population II stars have an [Fe/H] of less than -1.  There is a third category, known as Population III.  These would be the first stars of the universe, with essentially no “metals” in them.

We have yet to observe a Population III star, but this new discovery comes very close.  This new star, known as SM0313 contains no measurable traces of iron.  From its spectrum you can see carbon, calcium and magnesium, but nothing else beyond hydrogen.  In comparison to the spectrum of a typical low metallicity star the difference is striking.

Spectrum of SM0313. Credit: Anna Frebel.

The complete lack of measurable iron in the spectrum is surprising.  Based on the limits of their observations, the authors calculate the metallicity of SM0313 to be no more than Fe/H = -7.1.  This means it is likely a second generation star, and its first generation progenitor ended in a supernova that wasn’t powerful enough to cast out significant quantities of iron.

This changes our understanding of first generation stars. It has been thought that the large size of first generation stars meant their supernova would be a particularly powerful kind known as a pair-instability supernova. Such supernovae would cast out large quantities of material, and as a result the remnants of gas and dust would tend to be enriched and mixed on a relatively short cosmic time scale.  As a result, even second generation stars such as SM0313 would have some measurable levels of iron.  The discovery of SM0313 means that low energy first-generation supernovae were more common.  Thus the mixing and enrichment of gas and dust happened more gradually than originally expected.

Paper:  Keller SC, Bessell MS, Frebel A, et al. A single low-energy, iron-poor supernova as the source of metals in the star SMSS J031300.36-670839.3. Nature. 2014;

The post Second Generation appeared first on One Universe at a Time.