Iron-rich stars host planets on closer orbits than their iron-poor siblings, astronomers find. The results could help reveal how planets form.

Artist’s impression of the view just above the surface of one of the middle planets in the TRAPPIST-1 system. Impression based on the known physical parameters for the planets and stars seen, and using a vast database of objects in the Universe.
ESO/N. Bartmann/spaceengine.org

The more iron a star contains, the closer its planet’s orbit. And astronomers aren’t quite sure why.

Robert Wilson, a graduate student at the University of Virginia, announced the puzzling result at a meeting of the American Astronomical Society in Washington, D.C.

Stars are mostly hydrogen and helium, with just a smattering of heavier elements. Since stars forge heavy elements in their core, the ones we see on the surface come from previous generations of stars. The longer a star’s lineage, the more such elements enrich it (or pollute it, depending on your point of view). The heaviest element a star can make is iron, so its abundance serves as a proxy for the presence of all the other elements in the star, or in astro-speak, the star’s metallicity.

Planets form out of the same natal gas as their parent star. So a star’s high metallicity is a sign that its planets came together within metal-enriched gas. Previous studies have found that metallicity plays a role in planet formation — but astronomers don’t yet understand how the connection works.

Wilson studied metallicity’s effect on planet formation using data on 282 candidates discovered by the exoplanet-hunting Kepler mission. The Sloan 2.5-meter telescope in New Mexico took spectra of these systems as part of the APOGEE program, revealing each star’s iron abundance.

This artist's conception shows the silhouette of a rocky planet, dubbed HD 219134b, as it passes in front of its star.
NASA/JPL-Caltech

To Wilson’s surprise, the stars richest in iron host planets on scorchingly close orbits, while stars with lower iron abundances have planets on farther-out orbits. The results point to different formation histories for the two types of planets.

A clear line divides the two groups of planets: iron-rich stars host planets with orbits of 8 days or less, while the farther-out planets circle their iron-poor stars on periods longer than 8 days. Yet the two sets of stars aren’t all that different from each other — the ones labeled iron-rich have only 25% more iron than those labeled iron-poor.

“That’s like adding five-eighths of a teaspoon of salt into a cupcake recipe that calls for half a teaspoon, among all its other ingredients,” Wilson says. When baking a planet, it turns out, even a small difference in the metallicity of a planet’s natal cloud can have surprisingly strong effects on its formation.

But how? Wilson suspects that higher-metallicity gas makes for flatter planet-forming disks. The presence of heavy elements helps gas in the planetary disk cool and collapse to the centerline — like someone forgot the baking powder when making pancakes. Thinner disks make it easier for forming planets to migrate inward, closer to the star.

The next step will be an astronomer’s version of America’s Test Kitchen: Wilson is working with theorists to cook up stars and their planet-forming disks within different metallicity environments to see if they can reproduce the same iron-rich/iron-poor divide.

Comments


Image of

January 11, 2018 at 10:45 am

I am not a scientist, only an enthusiast, so my question may be a bit naïve & simplistic but here goes. Could the higher Iron content affect greater gravitational attraction or maybe have an effect on magnetic fields and solar winds?

You must be logged in to post a comment.

Image of

January 11, 2018 at 10:47 am

Me again; I forgot to ask: what is the Iron content of our Sun?

You must be logged in to post a comment.

Image of Rod

Rod

January 11, 2018 at 5:26 pm

Good question. The solar value is measured as log [FE/H] and solar value = 0 (or 1.0). A star with -2.0 is 0.01 solar value iron content where the sun is 1.0. A star with 2.0 is 100x the sun value. I did not see this in the report but the 25% iron difference does not seem like a large amount of difference in the stars sampled.

You must be logged in to post a comment.

Image of Rod

Rod

January 11, 2018 at 8:41 pm

John, this may help too. Iron is 0.014% of total solar mass, see https://www.space.com/17170-what-is-the-sun-made-of.html. Once you know the Fe/H value relative to the Sun, you can convert into mass using earth masses or grams or kilograms. The sun rounded is 2E+33 grams.

You must be logged in to post a comment.

Image of

January 12, 2018 at 7:37 pm

Thanks for the reply Roderick, and the link. It definitely ads new perspective for me.

You must be logged in to post a comment.

Image of cosmicrayray

cosmicrayray

January 13, 2018 at 1:16 pm

From another non-scientist: for the iron rich or iron "starved" bodies to be in different "orbits" does this not relate to Kepler's third law: either the mass, density, axis and period drives the situation? What do you think is driving this system where iron seems to be the driver? Other heavy atoms other than iron?

You must be logged in to post a comment.

Image of Rod

Rod

January 15, 2018 at 7:56 pm

cosmicrayray, interesting questions about different orbits. My opinion - the entire topic of hot gas in an accretion disk forming a variety of exoplanets by slowly growing from hot, microscopic particles to planetesimals and then on in size, remains more theory and many parameters adjusted in computer models. In observational astronomy, we have more than 3700 exoplanets documented now, most have no disk or dust disk. We can assume those disks were there initially, but this is also circular argument.

You must be logged in to post a comment.

Image of Harry

Harry

January 16, 2018 at 6:29 pm

I am wondering if the increase in Fe causes a delay in nuclear ignition of the star as it formed. Could gamma ray absorption effects of a higher mix of Fe delay nuclear ignition and the resultant stellar winds? I am not familiar with what Fe would do at those temperatures and pressure. With a delayed ignition, the disc of material would have more time to spiral in closer to the stellar surface before the nuclear winds and centrifugal momentum offset gravity.

You must be logged in to post a comment.

Image of Rod

Rod

January 17, 2018 at 2:04 pm

This report has generated some interesting discussion notes. Today I reviewed the exoplanet.eu catalog, now showing 3728 exoplanets. The average host star mass is 0.9948 solar masses and average host star metal content is log [FE/H] = -0.013867793. Converting to antilog and percent, that means the average iron content is about 97% solar compared to the Sun in the catalog. We are still talking about small amount of differences here concerning iron/hydrogen ratio. Perhaps more computer modeling of different accretion disks and dissipation/evaporation times will help.

You must be logged in to post a comment.

You must be logged in to post a comment.