Astrobites RSS

Proxima Centauri

Editor’s note: Astrobites is a graduate-student-run organization that digests astrophysical literature for undergraduate students. As part of the partnership between the AAS and astrobites, we occasionally repost astrobites content here at AAS Nova. We hope you enjoy this post from astrobites; the original can be viewed at astrobites.org!

Title: ALMA Discovery of Dust Belts Around Proxima Centauri
Authors: Guillem Anglada, Pedro J. Amado, Jose L. Ortiz, and collaborators
First Author’s Institution: Institute of Astrophysics of Andalusia, Spain
Status: Published in ApJL

Hiding behind the gleaming light of a pale red dot lies one of the world’s favorite exoplanets, Proxima b, whose orbital motion imprints a barely detectable wobble in its host star, Proxima Centauri. Following the seminal discovery of a planet in its habitable zone, this undistinguished low-mass red dwarf star became a target for tireless examination, despite the challenges involved in observing it. In today’s paper, we will see that one of these searches, however, seems to have paid off … in dust?

Figure 1. Observations of Proxima Centauri with the ALMA observatory suggest that it may not only host a rocky planet but also at least one dust ring around it. [Anglada et al. 2017]

The One Ring

Okay, I will be honest and admit that, indeed, few people in the world would be enthusiastic about dust — in space, of all things! But, let me tell you, this is exciting dust. Guillem Anglada and his collaborators (including his spacefaring self from a parallel universe, Guillem Anglada-Escudé, who is the first author of Proxima b’s discovery paper) used the Atacama Large Millimeter Array (ALMA) to observe a dust ring around Proxima Centauri. Such an object is a signpost of terrestrial planet formation and could clue us in on what its planetary system looks like and how it came to be.

You may recall that the lore of Middle Earth tells the story of The One Ring to Rule Them All™, whose (un)fortunate wearer is granted the power to become invisible. In a funny exchange of roles, the dust ring around Proxima Centauri may actually help us uncover more information about the well-hidden planet Proxima b, such as its orbital inclination and mass.

But leaving western epic novels aside, let’s talk about the instrument. The ALMA observatory has two types of antennas: the smaller 7-meter diameter detectors, and the bigger 12-meter ones. The data that comes from these antennas differ in that the smaller ones produce images with fewer details but with a wider field of view; the bigger antennas produce more detailed images but see a limited field of view.

Following analysis of the data depicted in Figure 2 below, the authors found that Proxima Centauri appears to emit more infrared light than it should, which they conclude must be produced by a “belt,” or ring, of cold (50 K) dust. The dust belt has a radius four times the distance of the Earth from the Sun (i.e. 4 AU). In reality, this dust ring seems to be an analog to the Solar System’s Kuiper Belt. They estimate that the mass contained in this belt is 1% that of Earth’s, which is similar to the mass of the Kuiper Belt around the Sun.

Figure 2. Image of Proxima Centauri (represented by the + mark) using ALMA’s 12-meter array. Although not clearly separated from the central source, the observed infrared excess strongly suggests the presence of a dust belt around the star with a radius smaller than 4 AU. Its shape also hints at the presence of a warmer belt closer to the central star, but this hypothesis needs confirmation. The identity of the detached turquoise blob to the left of the star is unknown: it could be either a real object or plain noise fluctuation (see main text). [Anglada et al. 2017]

Rings for Days

In addition to the “One Ring” described above, the elongated shape of the source in Figure 2 suggests the presence of a warmer (T ~ 90 K) dust belt with size 0.8 AU, but the authors don’t seem to be completely sure about its nature yet.

Now, if you’re asking yourself what that blob of emission to the lower left of Proxima Cen in Figure 2 could possibly be, the short answer is: we don’t know yet. The long answer is that it could either be a real source or it could just be random noise fluctuation. If it is a real source then the authors suggest many possible explanations, such as a background galaxy or a collision between large bodies. But the most intriguing explanation for this potential source is that we could be looking at the rings of a Saturn-like planet orbiting Proxima Cen. More observations will be needed to confirm this exciting possibility, though.

But wait, there’s more! Figure 3 below depicts the image as observed by the compact array configuration of ALMA, using the 7-meter antennas: what seems to be a second dust belt is seen as a series of green smudges at a distance of approximately 30 AU, represented by the white ellipse. Although the detection is very marginal, the authors propose that this could be an outer dust ring with 1/10,000 the mass of Earth.

Figure 3. Image of Proxima Centauri using ALMA’s compact array configuration (7-meter antennas). The dotted white ellipse marks the position of a dust belt of radius 30 AU. The green regions are the positions where the signal produced by the dust belt is stronger.

I know, I know, that was a lot to take in! Proxima Centauri suddenly seems a lot busier than we previously thought, likely sporting a rocky planet in its habitability zone and a Kuiper Belt-analog. The other possibilities, such as the Saturn-like ringed planet and the other dust belts are still a bit speculative, so further observations of the system are a no-brainer at this point.

About the author, Leonardo dos Santos:

Leo is an exoplanet scientist and Ph.D. candidate at the Geneva Observatory. His current research involves characterization of exoplanets, physical and chemical properties of stars similar to the Sun and developing astronomical software. Not to be confused with the constellation.

turbulent disk

Editor’s note: Astrobites is a graduate-student-run organization that digests astrophysical literature for undergraduate students. As part of the partnership between the AAS and astrobites, we occasionally repost astrobites content here at AAS Nova. We hope you enjoy this post from astrobites; the original can be viewed at astrobites.org!

Title: The Effects of Protostellar Disk Turbulence on CO Emission Lines: A Comparison Study of Disks With Constant CO Abundance vs. Chemically Evolving Disks
Authors: Mo (Emma) Yu, Neal Evans, Sarah Dodson-Robinson, Karen Willacy, Neal Turner
First Author’s Institution: University of Texas at Austin
Status: Accepted to ApJ

We know it happens. We see that protoplanetary disks — the birthplace of planets — spill the gaseous material at their inner edge onto the young stars around which they orbit. This process of accretion persists throughout the disk’s lifetime and within 1 to 5 million years (or 10 Myr in rare cases), a disk will feed all of its gas to its star and completely fade away — leaving behind only the planets that formed along the way and any leftover rocky material that remains.

The rate at which accretion occurs — as well as why accretion occurs — are both driving forces behind determining what types of planets can form quickly enough in a protoplanetary disk’s relatively short lifetime. While we can measure accretion rates onto stars directly, we still do not know why disks accrete! This is one of the most important unsolved problems in planet formation. Until we can solve it, our models for planet formation are incomplete.

There are two leading potential explanations for why accretion occurs: (1) turbulence, and (2) magnetic winds. In the last two years, several attempts have been made to measure turbulence in a nearby disk for the first time using CO (carbon monoxide) spectral lines. These measurements showed that the turbulence in that system is not strong enough to be responsible for accretion, winning favor for the other idea of magnetic winds.

However, today’s paper led by Emma Yu argues that those measurements were not interpreted properly since they did not take into account the relatively rapid rate at which CO depletes over time (see Figure 1). This paper asks: What can we really learn about turbulence from CO spectral lines if we include CO depletion in our model?

CO depletion

Figure 1: Density of CO and H2 at a range of distances from the star over time (taken from chemical models). Previous measurements of turbulence assumed the ratio of CO-to-H2 to be constant over time, but this is not true beyond 15 AU after 1 Myr. [Adapted from Yu et al. 2017]

Background: Why Do Protoplanetary Disks Accrete onto Their Stars?

It is understood that disks must accrete by transporting angular momentum outwards. Naively, we might expect this to happen through “viscous shear flow.” In this process, the gaseous disk struggles to behave like a fluid because it rotates at a slower rate further away from the star (due to Kepler’s 3rd law). As a result, material in the outer disk will try to speed up to catch up to the inner disk — thereby increasing its angular momentum. Meanwhile to compensate for that increase (and conserve angular momentum), the material in the inner disk will slow down. Since this gas is rotating too slowly to obey Kepler’s 3rd law, it will spiral inwards and eventually accrete.

There is just one problem: disks are too sparse to interact viscously like a fluid! For shear flow to occur, there must be some source of turbulence to make the disk viscous. It is widely accepted that magnetic fields can create turbulence by triggering an instability, but it is not clear if this instability actually occurs in the entire disk. If it does not occur in most of the disk, turbulence most likely would not be responsible for making the disk accrete.

Instead, magnetic winds — which transport angular momentum outward in a completely different way by flinging material out of the disk on magnetic field lines pointed away from the star while the disk rotates (see also here and here for more) — would be the favored reason for the disk accreting.

Introduction to Spectral Lines with Turbulence

We would know why protoplanetary disks accrete if we could just measure their levels of turbulence, but we did not have the telescope power to do this until the ALMA telescope array was turned on several years ago. In this astrobite, Tim discussed a new method for measuring turbulence with ALMA by looking at the shape of a specific CO spectral line.

Spectral lines are one of the most important tools in astronomy. They are used for everything from inferring compositions of atmospheres to measuring distances to distant galaxies, among many other things. Molecules (and atoms) emit spectral lines when electrons that were excited to a high energy state transition back to a lower energy state. While these lines are emitted at the precise wavelength corresponding to a transition, they never appear perfectly thin. They always exhibit some level of “broadening” due to the gas molecules moving towards us or away from us as they emit.

The velocities of individual molecules are mostly random. Thermal effects create most of the random motion for gas molecules. Besides that, turbulence also creates a little bit of random motion, causing spectral lines to broaden more than normal. As a result, we can use the broadened shapes of spectral lines (that are very well-resolved) to measure turbulence!

Spectral Lines with CO Depletion

The paper covered in Tim’s astrobite found that turbulence can be probed by measuring the CO spectral line’s peak-to-trough ratio (see Figure 2). Specifically, disks that are less turbulent have higher peak-to-trough ratios. However, this analysis assumed that the amount of CO in the disk relative to hydrogen is fixed over time, whereas chemical models predict it should actually drop significantly in older disks (see Figure 1).

CO spectral line profiles

Figure 2: CO spectral line profiles with different levels of turbulence, with lower levels producing a higher peak-to-trough ratio. (Note: The double-peak structure arises from part of the disk rotating towards us and part of the disk rotating away from us.) [Adapted from Yu et al. 2017]

Coincidentally, Emma Yu et al. find that the depletion of CO also creates spectral lines with higher peak-to-trough ratios (see Figure 3). As a result, the authors of the other study may have thought that they detected low levels of turbulence by measuring a high peak-to-trough ratio when they may really only be seeing evidence that CO has been depleted!

two disk stages

Figure 3: CO spectral line profiles at different stages in the disk lifetime. Left: Includes CO depletion. Right: Assumes constant CO-to-H2 ratio. With a constant CO ratio, the peak-to-trough ratio does not change after 1 Myr. With CO depletion, the peak-to-trough ratio increases over time just like with decreasing levels of turbulence. [Adapted from Yu et al. 2017]

Figures 2 and 3 (and also Figure 7 of the paper) show how easy it is to confuse a disk with no turbulence at all with a disk that has a moderate level of turbulence. If that extra turbulence is present, it might be strong enough to explain the observed levels of accretion in the previous study, giving favor to the long-assumed idea that turbulence drives accretion — not magnetic winds. However, the paper shows that our current measurements may not be able to distinguish these levels of turbulence, leaving the question of why disks accrete still unsolved.

Future Work

All hope is not lost! The authors point out that a different isotopologue of carbon monoxide (C18O) may be more useful for measuring turbulence. However, its spectral lines are weaker and thus, more difficult to resolve. They also point out that CO would be more helpful if we could also measure the precise level of CO depletion in a disk (rather than infer it from chemical models).

This is an exciting time for studying protoplanetary disks because we are finally beginning to scratch the surface of measuring turbulence. Getting proper measurements though, will require more digging.

About the author, Michael Hammer:

I am a 3rd-year graduate student at the University of Arizona, where I am working with Kaitlin Kratter on simulating planets, vortices, and other phenomena in protoplanetary disks. I am from Queens, NYC; but I’m not Spider-Man…

Aurorae on Jupiter

Editor’s note: Astrobites is a graduate-student-run organization that digests astrophysical literature for undergraduate students. As part of the partnership between the AAS and astrobites, we occasionally repost astrobites content here at AAS Nova. We hope you enjoy this post from astrobites; the original can be viewed at astrobites.org!

Title: The Detectability of Radio Auroral Emission from Proxima b
Authors: Blakesley Burkhart & Abraham Loeb
First Author’s Institution: Harvard-Smithsonian Center for Astrophysics
Status: Published in ApJL, open access

Dazzling auroral displays are not uncommon in our solar system. In fact, when the Sun sends highly energetic particles out in to the solar system, any planets with a substantial magnetic field will interact with the particles, resulting in the emission of radio waves. Detecting this emission allows us to determine many interesting planetary properties, such as orbital parameters, habitability, plate tectonics and atmospheric compositions. Yet we have not observed any auroral activity from planets that lie outside our solar system (we have, however, detected radio auroral emission on a brown dwarf star!).

Proxima b

Figure 1: Artist’s impression showing Proxima b orbiting its red dwarf host Proxima Centauri. [ESO/M.Kornmesser]

Today’s bite investigates our nearest stellar neighbour, Proxima Centauri, to answer one very important question: can we detect auroral emissions from its exoplanet, Proxima b?

What Are We Looking For?

To answer this question, the authors first employ radiometric Bode’s law to estimate the magnitude of radio waves released during stellar wind and magnetosphere interactions. Radiometric Bode’s law, derived from observations of magnetic planets within our own solar system, indicates that the brightness of radio waves increases with size of the planetary magnetosphere. Blackett’s law then suggests the size of the magnetosphere scales with mass and rotation speed, which translates to a predicted radio power of 1013 for Proxima b. When combined with Proxima b’s estimated magnetic field strengths of 0.007 – 1 G, comparable to the 0.5 G observed at the Earth’s equator, scientists expect the frequency of these radio waves to be between 0.02 – 2.8 MHz.

radio wave brightness

Figure 2: Predicted brightness of radio waves versus emission frequency, in accordance with Bode’s law, for 106 exoplanets. The expected values for Proxima b are highlighted in the blue square, with the yellow star representing a magnetic field value of 0.3 G. The black dashed line indicates the cut-off frequency for Jupiter. [Burkhart & Loeb 2017]

Figure 2 highlights the first hurdle for observations of radio auroral emissions from Proxima b. All radiometric modelling points to radio waves being emitted at frequencies between 0.02 – 2.8 MHz for this exoplanet. However observations below 10 MHz are not possible from Earth, as these wavelengths are blocked by our atmosphere — so even at the extremes of likely magnetic field strengths, we cannot use ground-based instrumentation. A secondary issue with observing aurorae on Proxima b is that we’re now observing in a low-frequency regime that tends to be absorbed by the interstellar medium (ISM).

Modelling a Magnetosphere

So far, the authors have only estimated the brightness of auroral activity on Proxima b. With its close-in orbit creating a highly variable magnetosphere radius, we expect the variation in the observed radio waves to be rather large. To characterise this variability, authors implemented models of the wind and magnetic field around Proxima b as a function of various orbital parameters. An example of the results obtained from the simulations is shown below in Figure 3.

Figure 3: Variation in the observed radio brightness as the radius of the magnetosphere caries over Proxima b’s short 11.2 day orbit. Eccentricity = 0 and inclination = 10 degrees. [Burkhart & Loeb 2017]

Here we see how the simulated brightness of radio waves changes over one orbit. The two dips correspond to periods where the magnetospheric radius suddenly changes as the planet passes through streamer regions, located near Proxima Centauri’s equator, where stellar winds become denser. The authors also considered four different magnetic field strengths and found that a weaker magnetic field, and therefore lower emission frequency, results in brighter radio emission.

From modelling the variability in radio flux for Proxima b, the authors concluded two things:

  1. Proxima b’s radio auroral emissions vary by almost an order of magnitude over one full orbit.
  2. The amplitude of the variation depends on orbital parameters — the eccentricity and inclination of the orbit — as well as on parameters of the stellar wind and planetary magnetic field.

Can We Detect Aurorae Around Proxima b?

There are a number of issues associated with undertaking observations of Proxima b’s radio emission. As previously mentioned, the ISM is highly problematic, as electrons within the ISM are more likely to absorb photons at low frequencies due to free-free absorption. Thankfully, this shouldn’t be an issue above 0.3 MHz for nearby planetary systems, like Proxima b, as there is less ISM to contend with. We also have to overcome the 10 MHz atmospheric cut-off introduced by absorption of photons in the ionosphere. Clearly the only solution here is to make observations from space. The authors mention several interesting proposals including a radio observatory on the Moon (one example given is ROLSS) and clusters of low-cost CubeSats to form a very large telescope (think of experiments like ALMA which combined lots of smaller telescopes to form one big one, but in space).

Figure 4: Example CubeSat with hands for scale. [NASA]

The take-home message of this paper is that the brightness of radio emissions around Proxima b are substantial enough to be detected here on Earth. This is fantastic because if aurorae, caused by interactions between stellar winds and the planet’s magnetosphere, are detected on Proxima b, we will be able to further constrain the planet’s orbital inclination, eccentricity and generally gain insight into its magnetosphere. Now we just have to wait for the right instrumentation to put into space, allowing scientists to overcome the pesky 10 MHz limit imposed by our own atmosphere.

About the author, Amber Hornsby:

First year postgraduate researcher based in the Astronomy Instrumentation Group at Cardiff University. Currently I am working on detectors for future observations of the Cosmic Microwave Background. Other interests include coffee, Star Trek and pizza.

NGC 1300

Editor’s note: Astrobites is a graduate-student-run organization that digests astrophysical literature for undergraduate students. As part of the partnership between the AAS and astrobites, we occasionally repost astrobites content here at AAS Nova. We hope you enjoy this post from astrobites; the original can be viewed at astrobites.org!

Title: Galaxy-Scale Bars in Late-Type Sloan Digital Sky Survey Galaxies Do Not Influence the Average Accretion Rates of Supermassive Black Holes
Authors: A.D. Goulding, E. Matthaey, J.E. Greene, et al.
First Author’s Institution: Princeton University
Status: Accepted to ApJ, open access

When it comes to picking their host galaxies, active galactic nuclei (or AGN) are rather promiscuous. They reside in all types of galaxies: ellipticals, irregulars, and spirals. AGN of the same feather tend to flock together — the more luminous and radio-loud ones are found in elliptical galaxies while the lower luminosity ones are more often found in spiral galaxies. This is a manifestation of the black hole mass-host galaxy luminosity correlation, where spiral galaxies like our Milky Way tend to have less massive black holes than elliptical galaxies. Besides spiral arms, spiral galaxies sometimes also boast of having bars, if the right mood strikes. How are bars related to their AGN? Could they trigger the central black holes to light up as AGN?

Galactic bars are thought to contribute to the dynamical evolution of their host galaxies. Numerical studies show that they can funnel in gas from the outskirts to the central regions of the galaxies, triggering star formation and possibly AGN activity. It is still unclear whether bars actually help trigger AGN, as previous studies have produced conflicting results and tend to suffer from small number statistics and biased AGN diagnostics. In today’s paper, the authors bring better tools to bear on the problem, by utilizing the large wealth of information from the SDSS Galaxy Zoo citizen science project and X-ray stacking analyses.

The authors first selected a sample of ~100,000 local spiral galaxies from SDSS that have been visually classified as such in Galaxy Zoo by at least 20 people. This sample is later divided into three redshift bins that are each complete in stellar mass, i.e. there are galaxies that span the whole range of stellar masses in each redshift bin. This is to ensure the final results are representative and unbiased. They distinguished sources with and without bars in these three redshift bins using information from Galaxy Zoo, based on the fraction of votes to the question “Is there a sign of a bar feature through the center of the galaxy?” If the fraction of votes is equal or greater than 0.25, the galaxy is defined to have a bar; if the fraction of votes is less than 0.1, the galaxy is defined to be bar-less. Galaxies with fraction of votes between these two numbers are defined to be ambiguous. Figure 1 shows some examples of spiral galaxies with bars, no bars, and ambiguous bars in their sample.

Fig. 1: Sample unbarred (blue borders), ambiguously barred (yellow borders), and barred (red borders) spiral galaxies from the Galaxy Zoo project, as determined by fbar, which is the fraction of votes by citizen scientists for the presence of bars. [Goulding et al. 2017]

In contrast to optical light which is absorbed by dust, X-rays from an AGN can more easily pierce through dust obscuration. Stacking lots of X-ray observations help to reveal heavily-obscured or low-luminosity AGN. Using data from the Chandra X-ray observatory, the authors performed X-ray stacking analyses to investigate the presence of AGN in their barred, unbarred, and ambiguously barred samples. Figure 2 shows the X-ray luminosity of their samples after subtracting the contributions from star formation processes. The three types of spiral galaxies do not show obvious differences in their X-ray luminosities, suggesting that AGN are no more common in one type of galaxy than the others. For galaxies with X-ray detections (i.e. hosting AGN), the authors further investigated the distributions of their specific black hole accretion rates, which are the X-ray luminosities divided by the host galaxy stellar mass shown in Figure 3. This ratio removes the dependence on stellar mass and instead probes the dependence on the host galaxy properties. There is again no difference in the accretion rates between the barred and unbarred samples.

Fig. 2: Star-formation subtracted X-ray luminosities vs. redshift in three X-ray energy bands. As with Figure 1, blue markers refer to unbarred galaxies, yellow to ambiguously barred, and red to barred galaxies. Open markers are sources with X-ray detections while filled markers are the luminosities produced by stacking sources without X-ray detections. As a comparison, the predicted mean X-ray luminosities due to stellar processes are shown by the dotted lines. [Adapted from Goulding et al. 2017]

Fig. 3: Distributions of specific black hole accretion rates for galaxies with AGN. The different line colors again refer to the presence or absence of bars. Dashed, solid, and dotted lines refer to different cuts in the X-ray luminosity. [Goulding et al. 2017]

Well, all that is a bummer — the presence of AGN in spirals seems to be independent of the presence of bars. For those with AGN, there is also no difference in the specific accretion rates of their host galaxies on the basis of a bar existence. As stacking analyses tend to wash away short timescale events, any bar contributions to AGN activities would need to be very short-lived. This study shows that over the lifetime of the galactic bars, they do not play significant role in triggering AGN — astronomers need to turn their eyes to other means of growing black holes in spiral galaxies.

About the author, Suk Sien Tie:

I am a third year PhD student at the Department of Astronomy at The Ohio State University. I am currently working on quantitative analyses of various quasar selection methods using the Dark Energy Survey (DES) and quasar variability via microlensing.

exoplanet atmosphere

Editor’s note: Astrobites is a graduate-student-run organization that digests astrophysical literature for undergraduate students. As part of the partnership between the AAS and astrobites, we occasionally repost astrobites content here at AAS Nova. We hope you enjoy this post from astrobites; the original can be viewed at astrobites.org!

Title: Redox Evolution via Gravitational Differentiation on Low Mass Planets: Implications for Biosignatures, Water Loss and Habitability
Authors: R. Wordsworth, L. Schaefer, R. Fischer
First Author’s Institution: Harvard University
Status: Submitted to ApJ, open access

Looking for Life

If you’ve been tuning into astronomy news lately, you’ve probably heard about a number of cool new exoplanet discoveries, like those in the TRAPPIST-1 system, continuously rolling in from our telescopes hard at work. But no matter how, when, and where a new exoplanet is discovered, there’s always that question burning at the back of our minds: could this exoplanet have Earth-like life?

Figure 1: A screenshot from the open-universe space exploration video game called No Man’s Sky. Even with Earth-like constraints, the sky’s the limit in terms of the many living worlds we can imagine are out there. [No Man’s Sky].

This question is certainly not an easy one to answer. For an exoplanet to house life (as we understand life so far), there’s a long checklist of requirements (like those discussed here and also here) that we need the exoplanet to fulfill. For example, life as we know it survives and thrives on liquid water, so we require that the exoplanet has the ability to hold liquid water.

Today’s astrobite focuses on another important requirement: the exoplanet’s atmospheric composition. Here on Earth, for instance, we have a lot of wonderful plant-based and plant-like creatures (like trees) that produce oxygen through photosynthesis. Then other creatures here (like humans) use that oxygen to survive and thrive. So for an exoplanet to have Earth-like life, we expect it to have a buildup of oxygen.

You might imagine, then, that we should be looking for oxygen on these exoplanets, as a sign of Earth-like life — and you’d be right, mostly! But unfortunately it’s not clear that a buildup of oxygen will “always” be a sign of Earth-like life. Instead, an exoplanet could possibly accumulate oxygen from purely chemical, completely not-organic-or-life-related (aka, abiotic) processes. This means that if we find an exoplanet that has a buildup of oxygen, we need to be cautious and somehow make sure that we haven’t just discovered a false positive (i.e., a case where what we think to be true is actually very false).

Today’s authors present a nifty theoretical framework for thinking about and modeling the atmospheres of exoplanets, which can help both characterize and predict the atmospheres of the many new exoplanets we’ve observed and continue to observe. The authors did many, many other cool things in their paper, but here we will focus on understanding the backbone of their theoretical framework.

A Little Chemistry

The authors use the concepts of ‘redox’ as the main variable of their model. ‘Redox’ is an abbreviation for ‘oxidation-reduction reaction‘, which is a fancy term for a reaction involving the exchange of electrons between two chemical species. The atom, molecule, or ion gaining an electron(s) is ‘reduced’, while the atom, molecule, or ion giving up an electron(s) is ‘oxidized’. Combustion is a common example of a redox. When you burn firewood at, say, a campfire, the carbon (C) from the wood reacts with the oxygen (O2) in the air, producing carbon dioxide (CO2). In this case, the carbon gives up four electrons and is oxidized, while the oxygen collectively gains four electrons and is reduced.

If a planet has an atmosphere that is very oxidizing, that means there is, in a sense, a net demand for electrons. That means we would expect O2, which wants to take electrons, to abiotically build up in an oxidizing atmosphere, because there would be no net supply of chemical species floating around for the O2 to react with and take electrons from. So to evaluate if an exoplanet would likely abiotically build up O2, we want to check if and when its atmosphere is likely to be oxidizing — and that depends on the planet’s chemical composition.

Figure 2 shows different elements from the periodic table as a function of electronegativity (which is basically how oxidizing versus reducing the elements are) plotted against atomic mass and abundance. With Earth as our example, we expect that the less massive elements like hydrogen (H), carbon (C), nitrogen (N), and oxygen (O) largely hang out in the volatile layer (which includes the atmosphere) and upwards towards space; magnesium (Mg), silicon (Si), and sulfur (S) largely get caught up in the planetary crust and mantle; and the heavy element iron (Fe) is largely trapped down in the core.

Figure 2: Electronegativity (x-axis) versus atomic mass (y-axis) for the most abundant elements in our solar system. Each circle corresponds to an element, which are labeled according to their periodic table abbreviations (H is hydrogen, for example). The sizes of the circles reflect the relative logarithmic abundances of these elements in our solar system. The faint gray asterisks in the background label elements less than 10% of the abundance of silicon (Si). Reducing elements are towards the left of the plot, while oxidizing elements are towards the right. And overall, the less massive elements rise towards space, while the more massive elements fall towards the planet’s core. [Wordsworth et al. 2017]

Studies have shown that, just like Earth, planets of around 1 to 10 Earth masses tend to split into these three major layers: volatile, mantle, and core. So to understand how oxidizing the atmosphere of a planet in this mass range is, we need to understand how these three planetary layers interact and exchange these different elements.

Putting It All Together

The authors quantify how oxidizing a planetary layer is in terms of the layer’s total oxidizing power using the following equation (Equation 1 in the paper):

Where N is the total oxidizing power, Ni is the number of atoms of some element ‘i’ (like carbon or oxygen), and pi is the oxidizing potential of element ‘i’. An atom of oxygen, for example, takes two electrons, so it has an oxidizing potential of +2. An atom of hydrogen, on the other hand, gives away one electron, so it has a negative oxidizing potential of -1. Finally, the Σ in the equation says that we want to sum over all elements ‘i’ in the layer. The authors drew out a model of the three-layer planetary system, as shown in Figure 3, and assigned a total oxidizing power N for each layer.

Figure 3: A three-layer model for a planet of about 1 to 10 Earth masses after formation. The N at the bottom-left of each layer represents that layer’s total oxidizing power. The k terms represent the exchange between these layers over time. And the E at the top represents the escape of the less-heavy elements, especially the reducing element hydrogen, out into space. [Wordsworth et al. 2017]

To calculate how oxidizing the atmosphere is with this model, we must track the flow of elements between these three layers over time. The authors discuss a lot of the cool processes that can allow transport of material between these layers at different times in the planet’s evolution. For example, when the planet is young, its mantle layer can be partially or fully ultra-hot and molten, like an ocean of magma. During this phase, the liquidy mantle layer and the atmosphere-containing volatile layer can interact and exchange materials much more easily. But once the planet cools down and the mantle layer hardens, the interactions get more complicated. Depending on the planet’s geodynamic structure (such as any churning plate tectonics and gas-spewing volcanoes), the mantle layer can shift around and still allow material to flow between these planetary layers. And over time the much-less massive elements, especially the reducing element of hydrogen, can leak out of a planet’s atmosphere and escape to space (that’s the ‘E’ term in Figure 2), also changing how oxidizing the atmosphere is.

Once we have a good handle on how these conditions and processes play out on other planets — such as answers to the tough question of how geodynamics evolve on planets other than Earth — we can apply this framework to dig out any planetary scenarios where compounds like O2 could build up abiotically over time. And for those seemingly habitable exoplanets that appear very unlikely to abiotically build up O2, we can be cautiously optimistic that any oxygen buildup we do see might come from a biotic source — exactly what we’ve been searching for all along.

About the author, Jamila Pegues:

Hi there! I’m a 2nd-year grad student at Harvard. I focus on the evolution of protoplanetary disks and extra-solar systems. I like using chemical/structural modeling and theory to explain what we see in observations. I’m also interested in artificial intelligence; I like trying to model processes of decision-making and utility with equations and algorithms. Outside of research, I enjoy running, cooking, reading stuff, and playing board/video games with friends. Fun Fact: I write trashy sci-fi novels! Stay tuned — maybe I’ll actually publish one someday!

PSR J1023+0038

Editor’s note: Astrobites is a graduate-student-run organization that digests astrophysical literature for undergraduate students. As part of the partnership between the AAS and astrobites, we occasionally repost astrobites content here at AAS Nova. We hope you enjoy this post from astrobites; the original can be viewed at astrobites.org!

Title: Simultaneous Chandra and VLA Observations of the Transitional Millisecond Pulsar PSR J1023+0038: Anti-correlated X-ray and Radio Variability
Authors: Slavko Bogdanov, Adam T. Deller, James C. A. Miller-Jones, et al.
First Author’s Institution: Columbia University
Status: Submitted to ApJ, open access

What’s more interesting than a rapidly spinning neutron star that emits electromagnetic radiation parallel to its magnetic poles? One that doesn’t exactly behave as expected, of course. One such weirdly acting pulsar, PSR J1023+0038, is a transitional millisecond pulsar (tMSP) — which is fancy speak for a pulsar with a millisecond rotational period that switches between radio and X-ray emission on a several-year timescale. The fact that this pulsar emits in both X-ray and radio on these longer timescales isn’t what piques the interest of astronomers, however, in the case of the study in this astrobite.

Weird Pulsar Behavior

Figure 1: Radio emissions (black) and x-ray emissions (blue) recorded by the VLA and Chandra respectively over time. This shows that when radio emissions drop off, X-ray emissions pick up.

Pulsars can typically fall into one of the following categories: radio pulsars are powered by exchanging rotational energy from the spinning neutron star into emitting radiation. This means that their rotation slows and their pulse length increases. Meanwhile, X-ray pulsars are accretion powered, meaning they turn heated infalling matter into X-ray emission. What distinguishes PSR J1023+0038 from the background of pulsars that switch between accretion-powered X-ray and rotation-powered radio pulsars is that it has a simultaneous anti-correlated X-ray and radio emission. The authors looked at about 5 hours of overlapping and concurrent observations from the Chandra X-ray Observatory and the Very Large Array (VLA) to try and understand this weird relationship between the X-ray and radio emissions. This is very clearly shown in Fig. 1 where we can see a tiny sample of time of overlapping X-ray and radio flux measurements. The anti-correlation is quite strong, meaning that when the X-ray emissions are weakest, the radio emission is strongest.

Figure 2: (Top) Chandra x-ray observations over a period of 5 hours.The 3 x-ray modes can be seen as the one large peak (~12.7 hrs), the low (minima), and high (steady maxima).    (Middle) The simultaneous observation as seen from the VLA. (Bottom) The overlapped top and middle observations show the anti-correlation between x-ray and radio emissions.

But wait…there’s more! When we zoom out on the flux/time series observations (Fig. 2) we can not only see that the anti-correlation is persistent, but we can also see that the X-ray emission has at least 3 unique modes of operation. The authors classify these X-ray emission modes as (1) sporadic flaring (~12.7 hrs in Fig. 2), (2) high, and (3) low modes. The difference between high and low in this context is the magnitude of the luminosity.

Trying to Explain Away the Strangeness

This complex and weird behavior unfortunately does not come with an easy or readily available answer. What we know about pulsars and how we can model pulsar accretion doesn’t shed any new light on the situation. The authors do suggest that the switching between high and low modes might occur due to a changing unstable magnetosphere. They also propose that the increase in radio emission can be explained by an outflow of plasma that emits synchrotron radiation as it travels. Additionally, when comparing PSR J1023+0038 to a low-mass X-ray-emitting binary black hole (BH LMXB) (Fig. 3), we can see that the low mode of this tMSP falls into the binary BH region. This is unusual because there is a pretty clear separation between the X-ray/radio luminosity relationship of neutron stars and BHs. Knowing this now, it may call into question whether some BH LMXBs could have been misidentified.

Figure 3: PSR J1023+0038 low (red diamond) and high (red pentagon) X-ray modes closely follow the accreting low-mass X-ray binary black hole systems, making it indistinguishable in the low mode.

Now you may be asking, “so what did we actually discover?”, which is a completely valid question. Well for one, we learned that there in fact do exist strange and unique pulsars that exhibit odd behavior. But the more exciting result is that we may not have a great understanding of pulsars in general. This is exciting because it can spur new astrophysical theories and models; ones that can more generally explain even the weirdest behaviors. Like most of astronomy (and science in general) however, before we can fully claim any specific mechanism for causing the anti-correlated X-ray and radio emissions and the switching between emission modes, we’ll probably need more observations.

About the author, Joshua Kerrigan:

I’m a 3rd Year PhD student at Brown University studying the early universe through the 21cm neutral hydrogen emission. I do this by using radio interferometer arrays such as the Precision Array for Probing the Epoch of Reionization (PAPER) and the Hydrogen Epoch of Reionization Array (HERA).

Hyades cluster

Editor’s note: Astrobites is a graduate-student-run organization that digests astrophysical literature for undergraduate students. As part of the partnership between the AAS and astrobites, we repost astrobites content here at AAS Nova once a week. We hope you enjoy this post from astrobites; the original can be viewed at astrobites.org!

Title: K2-NnnA b: A Binary System In The Hyades Cluster Hosting A Neptune-Sized Planet
Authors: David R. Ciardi, Ian J. M. Crossfield, Adina D. Feinstein et al.
First Author’s Institution: Caltech/IPAC-NASA Exoplanet Science Institute
Status: Submitted to AJ, open access

The authors of today’s paper report the discovery of a Neptune-sized planet in the nearby Hyades Cluster orbiting within a binary star system. A binary consists of two stars mutually orbiting their center of gravity; the larger star is called the primary and the smaller is the secondary. In a close binary where the stars are separated only by a short distance, each of the two stars affects the evolution and size of the other.

The binary system in today’s paper is referred to as EPIC 247589423 (also called LP 358-348), but the primary and secondary stars within this system are called K2-NnnA and K2-NnnB, respectively. Since the planet discovered in this system orbits just the primary star, the authors of this paper refer to it as K2-NnnA b: exoplanets are generally named after their host star and then have another letter added to the end. However, this planet does not yet officially have a name, so there is hope that it will be called something that rolls off the tongue a bit more easily.

Significance to Planetary Formation and Evolution

Planets in binaries form under the influence of two stars — an extreme environment for planet formation, particularly if the two stars are close. The system EPIC 247589423 has a very close separation of 40 AU, which is roughly the distance from Pluto to the Sun. Using our solar system as an example, if we replaced Pluto with a star, it is easy to see that the rotation and generally everything about the planets would be drastically affected.

Studying binary-orbiting planets like K2-NnnA b allows us to test the robustness of planet formation, since the planets have to survive in an extreme environment for a long time. These systems also let us test how often planets are retained by their host star rather than being destroyed by the changing gravitational field. What’s more, by finding and studying planets in star clusters, we may begin to understand how planetary systems form and evolve and find out the timescale for such events.

K2 and Follow-up Observations

K2, or as I like to call it, Zombie Kepler, is the current mission for the Kepler Space Telescope. The Kepler Space Telescope was launched in 2009 with the mission of finding exoplanets through transit photometry. The method of transit photometry means recording the brightness of the star for an extended period of time, then looking for any small periodic ‘dips’ in the brightness that occur when an exoplanet eclipses a small part of the star. This data is generally referred to as a light curve, since there is a rounded dip where the exoplanet eclipses the star. After a mechanical mishap in 2013 with the Kepler Space Telescope, the scientists working on this mission still found a way to use the telescope to look for transiting exoplanets. Instead of observing one single part of the sky for an extended period of time as before, K2 now observes smaller patches of the sky for shorter amounts of time. Figure 1 shows several stages of the light curve analysis from K2.

Figure 1: This shows the light curve of EPIC 247589423 in various stages of analysis from K2. The topmost panel shows the light curve with the telescope rotation removed. The second panel shows the binned version of the top panel. The third shows the data with the stellar variability removed, and the lowest panel shows the folded and binned result for the planet transit. [Ciardi et al. 2017]

Scientists discovered the short-period exoplanet in the EPIC 247589423 system by examining the star’s K2-obtained light curve. Since K2-NnnA b orbits a binary system, the researchers had to first remove the stellar variability inherent in the data. The two stars within the binary already eclipse each other and cause dips in the light curve, which can make it difficult to find a much smaller dip from the planet. After removing the eclipse variability, they found the planet and solved for its period and radius.

After the initial discovery with K2, the researchers followed up their detection with new observations and archival data to confirm the planet’s presence. They used the archival data from 1950 Palomar Observatory Sky Survey to rule out any other object that could be causing the dip in the light curve — for example, another eclipsing binary behind the system, which could create variation in the light curve that could be mistaken for a planet — and then made additional observations of the system using the Keck I telescope.

K2-nnnA b: Planet Properties

The EPIC 247589423 binary system is located in the nearby Hyades cluster — a cluster roughly 750 Myr old and the nearest star cluster to the Sun. The two stars are separated by about 40 AU. The planet orbits the star K2-NnnA with a period of 17.3 days, and its transit lasts roughly 3 hours.

K2-NnnA b is one of the first Neptune-sized planets that has been observed orbiting in a binary system within an open cluster. The discovery of this planet can provide us with a better understanding of the planet population in stellar clusters and allow us to place more limits on planetary formation and evolution.

The authors of this paper say planets discovered in nearby star clusters ‘provide snapshots in time and represent the first steps in mapping out [planetary]evolution,’ and I wholeheartedly agree. The discovery of K2-NnnA b brings with it new understanding of planet formation in star clusters and in binary systems. The possibilities of planet formation and evolution are certainly not limitless, and with more and more discoveries like K2-NnnA b, we can hopefully find the extremes of planetary systems.

About the author, Mara Zimmerman:

Mara is working on her PhD in astronomy at the University of Wyoming. She has done research with Heartbeat binary stars and currently works on modeling debris disks.

S0-2

Editor’s note: Astrobites is a graduate-student-run organization that digests astrophysical literature for undergraduate students. As part of the partnership between the AAS and astrobites, we repost astrobites content here at AAS Nova once a week. We hope you enjoy this post from astrobites; the original can be viewed at astrobites.org!

Title: Investigating the Binarity of S0-2: Implications for its Origins and Robustness as a Probe of the Laws of Gravity around a Supermassive Black Hole
Authors: D. S. Chu, T. Do, A. Hees, A. Ghez, et al.
First Author’s Institution: University of California, Los Angeles
Status: Submitted to ApJ, open access

The most exciting discoveries in astronomy all have something in common: they let us marvel at the fact that nature obeys laws of physics. The star S0-2 is one of these exciting discoveries. S0-2 (also known as S2) is a fast-moving star that has been observed to follow a full elliptical, 16-year orbit around the Milky Way’s central supermassive black hole, precisely according to Kepler’s laws of planetary motion. Serving as a test-particle probe of the gravitational potential, S0-2 provides some of the best constraints on the black hole’s mass and distance yet. S0-2 is the brightest of the S-stars, a group of young main-sequence stars concentrated within the inner 1” (0.13 ly) of the nuclear star cluster.

The next time S0-2 reaches its closest approach to the black hole, in 2018, there will exist a unique opportunity to detect a deviation from Keplerian motion — namely the relativistic redshift of S0-2’s radial (line-of-sight) velocity — in a direct measurement. In anticipation of this event, the authors of today’s paper investigate possible consequences of S0-2 being not a single star, but a spectroscopic binary, which would complicate this measurement.

Figure 1: Top: Radial velocity measurements of S0-2 over time. Bottom: Residual velocities after subtraction of the best-fit model for the orbital motion. [Chu et al. 2017]

To search for any periodicity in S0-2’s radial velocity curve that would indicate the presence of a companion star, the authors combine their most recent velocity measurements with previous ones obtained as part of monitoring programs carried out at both the WMKO in Hawaii and the VLT in Chile. The resulting data set consists of 87 measurements in total, which are spread over 17 years of observations and have a typical uncertainty of a few 10 km/s (Figure 1, top panel). When S0-2 passes the black hole, the relativistic redshift of its radial velocity is predicted to amount to roughly 200 km/s at closest approach, while the radial velocity is expected to change from +4000 to -2000 km/s. S0-2’s actual speed at this time will be close to 8000 km/s, about 2.7% of the speed of light.

Figure 2: Lomb-Scargle periodogram of S0-2’s residual radial velocity curve (see Fig. 1). No peak reaches the 95% confidence level, which implies that no statistically significant signature of a periodic variation is found in the observations. [Chu et al. 2017]

After having accounted for the long-term radial velocity variation due to the orbital motion of S0-2 (Figure 1, bottom panel), the authors create a Lomb-Scargle periodogram to search for short-term periodic signatures in the velocity residuals. A companion of S0-2 would need to have an orbital period shorter than 120 days at maximum, or the binary system would be too wide to remain stable against tidal forces so close to the black hole. The minimum orbital period could be no less than a few days, or the two stars would come into contact. Yet even in between these limits, the measured periodogram shows no statistically significant peak at any particular period (Figure 2).

However, this non-detection of a periodic signal places an upper limit on the radial velocity variations that could be caused by a possible companion of S0-2, which can be converted into a mass limit. For instance, at a period of 100 days, velocity changes larger than about 12 km/s would have been confidently detected. This implies a companion mass smaller than about 1.7 solar masses, assuming a reasonable total mass of the binary in the range of 14.1 to 20 solar masses.

To estimate the effect of such a companion on the prospective measurement of the relativistic redshift, the authors simulate observations of S0-2 extending into 2018, using a relativistic orbit model and assuming that S0-2 is in fact a binary. These data sets are then fit in the same way as the real data would normally be, using a model in which S0-2 is assumed to be a single star and the strength of the expected relativistic effect is described by a free parameter. The authors conclude that even if S0-2 is a binary, the relativistic redshift could still be detected with high statistical significance in 2018, although the measurement could come out slightly biased, depending on the specific configuration of the binary system (Figure 3).

Figure 3: The bias in estimating the parameter describing the strength of the relativistic redshift, if S0-2 is a binary but assumed to be a single star, for different realizations of plausible binary configurations. Shown are deviations from the expected value of this parameter in general relativity, which is 1, where for a Keplerian orbit it is 0. [Chu et al. 2017]

A continued monitoring beyond 2018 will provide further opportunities to detect relativistic effects on the on-sky motion of S0-2 as well, and it remains to be studied how a possible binarity would influence those particular measurements. The authors also note that if, in time, the search for spectroscopic binaries could be extended to the fainter S-stars too, a comprehensive study of their binary fraction should be able to distinguish between different proposed scenarios for their formation. So stay tuned!

About the author, Philipp Plewa:

I am currently a graduate student at the Max-Planck-Institute for Extraterrestrial Physics in Germany. My main interest is in developing new tools for high-precision infrared astrometry, with the aim of learning more about the supermassive black hole at the center of the Milky Way and the stars in its vicinity.

Illustris project

Editor’s note: Astrobites is a graduate-student-run organization that digests astrophysical literature for undergraduate students. As part of the partnership between the AAS and astrobites, we repost astrobites content here at AAS Nova once a week. We hope you enjoy this post from astrobites; the original can be viewed at astrobites.org!

Title: Log-Normal Star Formation Histories in Simulated and Observed Galaxies
Authors: Benedikt Diemer, Martin Sparre, Louis E. Abramson, and Paul Torrey
First Author’s Institution: Harvard-Smithsonian Center for Astrophysics
Status: Published in ApJ, open access

Every galaxy has a story to tell. And every story has a few common plot devices: violent supernovae from dying young stars, bursts of activity from central supermassive black holes, mergers with other galaxies, and other dramatic astrophysical events that can change the course of a galaxy’s evolution. One galaxy property in particular can be severely impacted by these events, and that’s how many stars it is forming at a given time. The star formation rate (SFR) throughout the lifetime of a galaxy is known as its star formation history (SFH), and it is of interest to physicists trying to understand the lives of all galaxies.

Unfortunately, we can only see the stars that are in the galaxy at any given time; to infer the past history of star formation we must look for evidence of previous star-forming events. For nearby galaxies, we can view enough detail to see separate populations of stars and determine their individual ages. But for galaxies further away we can only observe the combined light of all the stars mixed together. Then we have to disentangle each component in order to get the underlying history of star formation.

However, in simulations of galaxies we can explicitly see the SFH in high resolution (see these ‘bites for previous examples). The authors of today’s paper looked at galaxies in the state-of-the-art Illustris simulation. They found that, whilst SFHs tend to be noisy, on average they show similar shapes over time. This shape is known as the log-normal distribution, and it typically exhibits a sharp rise to a peak, then a gradual fall.

Example Star Formation Histories

Figure 1: Top panels show the SFHs of two different galaxies in Illustris, measured in solar masses per year. The bottom panels show the cumulative SFR in solar masses. Galaxy (a) is a massive central galaxy, with a thousand billion Suns’ worth of stars, whereas galaxy (b) is a smaller satellite galaxy whose gas gets stripped by its host, ending up with only a billion solar masses of stars.

Figure 2 shows another two galaxies that are also very different. Both are still forming stars today, and galaxy (d) actually has an increasing SFR. Despite having very different forms for their SFH, the log-normal still provides a good fit.

Star Formation Histories

Figure 2: As for figure 1, but for two late-forming galaxies. Galaxy (c) is a galaxy that is still forming stars today, and (d) is a galaxy whose SFR is actually rising.

The authors find a correlation between the time of the peak and the width of the distribution: earlier peaks tend to be narrower, whilst later peaks tend to be much wider. In other words, galaxies that form early assemble quickly, whereas galaxies that form later take their time, leisurely building up mass over a longer period. You can see this in the examples in figures 1 and 2; the galaxies that form most of their stars early have narrow distributions, whereas those that are still forming stars have much wider distributions.

Log-normals are not without their limitations. Interactions between galaxies, such as mergers, can lead to bursts followed by sudden shutdowns of star formation that log-normals struggle to fit. But for most massive galaxies, log-normals describe their story arc in terms of star formation very, very well, helping physicists to understand every galaxy’s story, from start to finish.

About the author, Christopher Lovell:

I’m a 2nd year postgrad at the University of Sussex. I model high redshift galaxies using hydrodynamical simulations. When I’m not reading for work I read for pleasure, mostly science fiction and history, and when I’m not reading I enjoy dodging London traffic on my bike.

white dwarf

Editor’s note: Astrobites is a graduate-student-run organization that digests astrophysical literature for undergraduate students. As part of the partnership between the AAS and astrobites, we repost astrobites content here at AAS Nova once a week. We hope you enjoy this post from astrobites; the original can be viewed at astrobites.org!

Title: Astrophysical Implications of a New Dynamical Mass for the Nearby White Dwarf 40 Eridani B
Authors: Howard E. Bond, P. Bergeron, A. Bedard
First Author’s Institution: Pennsylvania State University
Status: Accepted to ApJ, open access

About a hundred years ago, one particular star made many astronomers scratch their heads. The star in question is 40 Eridani B, which is part of triple stellar system in the Sun’s neighbourhood, only 5.0 parsec away from us. It was classified as an A-type star by Williamina Fleming, one of the women computers working at the Harvard College Observatory. Fleming devised a system to classify stars according to the relative amount of hydrogen observed in their spectra, which was later improved by Annie Jump Cannon and is the basis of the system still in use. A-type, in particular, means that the star shows hydrogen as its most abundant element. This is usually a consequence of the star being so hot that molecules can’t stick together and metals are mostly ionised. As luminosity depends on temperature, this type of star is also very bright. 40 Eridani B, however, was too faint for a typical A star. A few years later, another faint star was also found to be of A-type: Sirius B, the companion to Sirius A, the brightest star in the sky. This suggested the existence of a new class of underluminous stars with spectra dominated by hydrogen.

You might have already guessed what they are, but it took astronomers back in the day a while to figure it out. Astronomer Willem Luyten was the first to refer to them as “white dwarfs”, back in 1922. These stars have already retired from the job of synthesising new elements, so there’s nothing to prevent gravity from acting and compressing them to a very small (about the size of Earth), extremely dense (a teaspoon of its material would weigh a tonne!) object. Unlike the typical A stars, the reason we can only detect hydrogen in their atmospheres is not their temperature, but the fact that their gravity is so strong that it pulls all the heavier elements to the core. As a result, they can be dominated by hydrogen and still be faint; thus, the mystery of the faint star was solved.

The Mystery of Core Composition

As our methods of measurement improved, astronomers were able to model the orbit of the stars in the system and use Kepler’s laws to constrain their mass. In 1974, astronomer W. D. Heintz obtained a mass of 0.43 times the mass of the Sun for 40 Eridani B. At the time astronomers already knew, however, that the lowest possible mass of a white dwarf that can be formed through single-star evolution — considering the present age of the Universe — is about 0.5 times the mass of the Sun. Any star that would form white dwarfs lighter than that should still be in the main sequence, calmly burning hydrogen into helium. So assuming the mass for 40 Eridani B was correct, it would have to be a helium-core white dwarf resulting from a binary-star merger. In this situation, the stars release energy when spiralling into each other, and this energy carries away material from the outer layers, allowing the formation of a lower mass white dwarf with a helium core.

A new mystery surrounding 40 Eridani B arose in 1998 when J. L. Provencal and collaborators estimated the radius of the star using parallax measurements. With the independent estimates of mass and radius for this and other stars, they proceeded to test the mass-radius relationship for white dwarfs (you can read more about it in this astrobite). They found that the combination of mass and radius of 40 Eridani B would be better explained by a core composed of a combination of magnesium and iron. So our theory of stellar evolution suggested the star had a helium core, while our theory of white-dwarf structure indicated a core composed of much heavier elements.

The Mystery of Heavier Mass

The plot thickened in 2012 when N. Giammichele and collaborators estimated the mass of 40 Eridani B using a model atmosphere analysis. This consists of comparing the spectrum of the star to model spectra derived from atmospheric models to obtain temperature and gravity, and subsequently estimating the mass of the star adopting a theoretical mass-radius relationship. This is by far the most widely used method to estimate the mass of white dwarfs, so we expect it to be reliable. Hence astronomers were very surprised to find that this method yielded a mass of 0.59 solar masses for 40 Eridani B, inconsistent with the mass derived from the orbital analysis. Now our theory of stellar evolution seemed to be clashing with our theory of stellar atmospheres as well.

Which theories should be revised? The authors of today’s paper have the answer.

More Data!

The authors were very surprised to discover that the orbit of 40 Eridani B had not been updated with new observations over the more than four decades since the work of Heintz. They contacted the binary star group at the United States Naval Observatory (USNO), which in turn assembled all the data they could find containing this star. With these new measurements, the group found a smaller orbital period, which led them to obtain a much higher mass for 40 Eridani B than Heintz: 0.573 solar masses.

This new mass agrees within uncertainties with the value derived by Giammichele and collaborators. To double check, the authors also fitted all the available spectra of 40 Eridani B to obtain the average gravity and, using a radius derived from parallax measurements, calculated a new mass. They found a mass of 0.565 solar masses, which, taking the uncertainties into account, also agrees with the new value derived from the orbit. We can thus rest assured that our spectral analysis method for deriving masses indeed works! The mystery of the heavier mass is solved.
Finally, the authors compared the new independent estimates of mass and radius to theoretical mass-radius relationships. As you can see in Fig. 1, the mystery of the core composition is also solved. As the mass was revised above the limit for single-star evolution, we would expect 40 Eridani B to be a carbon-oxygen core white dwarf. Those two elements are the heaviest mosts stars can synthesise, so they form the core of most white dwarfs we know. This indeed turns out to be the case for 40 Eridani B. The only catch is that the authors have to assume a thin hydrogen atmosphere, while our modern theories of stellar evolution mostly predict white dwarfs to have thick hydrogen layers.

Figure 1: The position of 40 Eridani B in a mass-radius plane given the new derived values for mass and radius is marked by the blue circle. The orange line shows the mass-radius relationship for an iron core, which was needed to explain previous estimates of mass and radius, but is very far from the new ones. The blue lines show the relationship assuming a carbon-oxygen core. The dash-dot line is for a thick hydrogen atmosphere, while the solid line, which agrees better with the data, assumes a thin hydrogen atmosphere. [Bond et al. 2017]

You usually read on Astrobites that we need more data to reach conclusions. Today’s paper is a nice example of how this actually works. Analysing more data, the authors proved that both the mass and the core composition of 40 Eridani B are well explained by our stellar evolution theory. They have also confirmed that the popular method of combining spectral analysis and theoretical mass-radius relationship to estimate stellar mass works quite well. None of those needs to be revised thus far. But luckily our work doesn’t end there: we have now to scratch our heads to figure out why the atmosphere of 40 Eridani B is thinner than our theory predicts. As often happens in science, solving one mystery ended up raising another.

About the author, Ingrid Pelisoli:

I am a third year PhD student at Universidade Federal do Rio Grande do Sul, in Brazil, and currently a visiting academic at the University of Warwick, UK. I study white dwarf stars and (try to) use what we learn about them to understand more about the structure and evolution of our galaxy. When I am not sciencing, I like to binge-watch sci-fi and fantasy series, eat pizza, and drink beer.

1 … 32 33 34 35 36 … 40