Exoplanet formation scenarios

Here’s a nice graphic by Sean Raymond illustrating different scenarios for the formation of exoplanetary systems, one leading to “Super-Earths” and the other to gas giants. The work is explained more fully on arXiv.

The paper’s figure caption includes:

Left: Evolution of the “breaking the chains” migration model for the origin of super-Earths. Embryos within the snow line are entirely rocky and much smaller than those that form past the snow line, which also incorporate ice. Presumably ice-rich embryos migrate inward through the rocky material, catalyzing the growth of purely rocky planets interior to the ice-rich ones. Planets migrate into long chains of mean motion resonances, with the innermost planet at the inner edge of the disk. The vast majority (90–95%) of resonant chains become unstable when the gas disk dissipates. The resulting planets match the distributions of known super-Earths.

Right: Evolution of the planet-planet scattering model for the origin of giant exoplanets. Several embryos grow quickly enough to accrete gas and grow into gas giants. They subsequently migrate into a resonant chain without drastically affecting the orbits of nearby growing rocky planets (or outer planetesimal disks). After the disk dissipates, the vast majority (75–90%) of giant planets systems become unstable. The resulting systems match the correlated mass-eccentricity distribution of known giant exoplanets.

Solar System planet tilts

Planetary scientist James O’Donoghue has put together a neat animation of the tilts and spin rates of the Solar System planets. Click on his Tweet to see the animation, or see the the higher-resolution version here.

The animation shows graphically that the simplistic idea that planets form in an orderly fashion from a proto-planetary disc cannot be entirely right. Uranus is tilted over; Venus spins backwards, rather slowly. Why? There must have been collisions, near collisions, and planets perturbing the orbits of other planets early on in our Solar System’s history. Thus we expect that the same thing will have occurred in the exoplanetary systems that we are now finding.

WASP-134b and WASP-134c: a pair of warm Jupiters

Most of the planets that WASP discovers are “hot Jupiters”, often defined as having an orbital period less than 10 days, though they clump at periods of 3 to 5 days. Occasionally we find “warm Jupiters”, with periods greater than 10 days. There seem to be far fewer of these (and not just because they’re harder to find, which they are, owing to being less likely to transit, because they are further away, and because they produce fewer transits because of the longer periods).

Our latest discovery paper, led by David Anderson, announces the WASP-134 system. An analysis of the radial-velocity observations looks like this:

There are clearly two different cycles from two different planets. Both are warm Jupiters. The inner one (upper panel) has a period just over 10 days while the outer one (lower panel) has a 70-day period. Both orbits are eccentric (the fits are clearly not sinusoids) and both planets have a mass of about one Jupiter.

This is relatively rare. Few systems are known where a shorter-period, Jupiter-mass planet has a Jupiter-mass companion with an orbit as short as 70 days. (Several systems are known where the companion is much further out, with a period of hundreds of days.)

The presence of two such planets makes it unlikely that the inner one got to its present position by the Lidov–Kozai “high eccentricity migration” pathways that are thought to explain many hot Jupiters. Such a pathway for one planet would be disrupted by the presence of the second planet.

This means that it is more likely that the two planets, WASP-134b and WASP-134c, either formed where they are, or moved inwards by “disc migration” mechanisms. Thus the two WASP-134 planets are perhaps a different population, with a different past history, than the majority of the planets found by WASP.

Sapphires and Rubies in the Sky

The Universities of Cambridge and Zurich have put out a press release about a study led by Caroline Dorn. The work discusses how planets form out of proto-planetary discs, and proposes that some planets would form at high temperatures out of condensates rich in Calcium and Aluminium. Their cores could thus effectively be giant rubies or sapphires (different forms of aluminium oxide).

Planets forming at different distances from their star will form at different temperatures, where different minerals will condense out.

Dorn et al suggest that the three planets HD219134 b, 55 Cancri e and WASP-47 e likely to be such objects. “In our calculations we found that these planets have 10-20% lower densities than the Earth”, says Caroline Dorn. The authors suggest that this is because they are rich in Calcium and Aluminium whereas other rocky planets have Iron-rich cores.

A depiction of 55 Cnc e (credit: Thibaut Roger)

“So, we have found three candidates that belong to a new class of super-Earths with this exotic composition,” says Dorn, adding that: “What is exciting is that these objects are completely different from the majority of Earth-like planets, if they actually exist.”

Take up of the press release has included the International Business Times, India Today, Popular Science, First Post, Sputnik News, ZME Science and other websites.

Spectral contamination from starspots on WASP-4

Here’s a topic we’ll be hearing much more about: how the observed spectrum of a transiting exoplanet is affected by transiting across star-spots. In “transmission spectroscopy” the starlight shines through the planet’s atmosphere during transit, and the easiest thing to do is assume that the star itself is a uniform light source.

But as discussed by papers led by Ben Rackham, if the planet passes over a dark region (star spot) or bright region (faculae), this would change the observed spectrum.

A new paper led by Alex Bixel about WASP-4b is the first to attempt to correct for this effect. The authors’ transit observations show a clear crossing of a starspot (the feature is shown in blue, the spot shows as a upward bump since the planet is then removing less light):

And here is the difference it makes. The blue curve is the observed spectrum, presumed to be of the planet’s atmosphere. The orange curve is then the spectrum corrected for the presence of the star spot.

The details of how to do this are complex, and are discussed at length in the above papers. The central message is that “active FGK host stars can produce such features and care is warranted in interpreting transmission spectra from these systems”.

However, there is good news in that: “stellar contamination in transmission spectra of FGK-hosted exoplanets is generally less problematic than for exoplanets orbiting M dwarfs”, and that such signals “are generally minor at wavelengths of planetary atomic and molecular features”. Overall the authors say that their study “bodes well for high-precision observations of these targets”.

Helium in WASP-69b, HAT-P-11b and HD 189733b

Earlier this year helium was found in the outer atmosphere of WASP-107b, the first detection of helium in an exoplanet. Several teams have now used similar techniques to find helium in WASP-69b, HAT-P-11b and HD 189733b, leading to a slew of papers and accompanying press releases from the Instituto de Astrofísica de Andalucía, the University of Exeter and others (see [1], [2], [3] and [4]).

Artist’s impression of an escaping envelope of helium surrounding WASP-69b. (Credit: Gabriel Perez Diaz, IAC)

Lisa Nortmann, lead author of the WASP-69b paper, explains that the helium is escaping from the atmosphere, forming a comet-like tail: “We observed a stronger and longer-lasting dimming of the starlight in a region of the spectrum where helium gas absorbs light. The longer duration of this absorption allows us to infer the presence of a tail.”

The press releases have led to extensive coverage including by CNN, the Daily Mail and Tech Times.

The IAA press release includes a video illustration of WASP-69b, created by Gabriel Perez Diaz of the IAC:

Is WASP-12b’s orbital decay driven by obliquity tides?

Tidal interactions between hot-Jupiter exoplanets and the host star should be causing their orbits to decay, such that the planet gradually spirals inwards. For most systems the change would be too small to detect in the decade or so that we’ve been observing them. However, WASP-12b is an exception, showing a clear change in its orbital period.

In a new paper on arXiv, Gracjan Maciejewski et al present the latest data for WASP-12b:

The graph records the change in transit time (“observed minus calculated” times, or O–C), showing that the transits are now occurring eight minutes early owing to a decreasing orbital period.

Such a rate is far faster than observed in other systems, and too large to be explained by the standard theory of tidal interactions.

However, a new paper led by Sarah Millholland suggests an answer. She suggests that the planet is tilted over, so that the axis around which it spins is tilted with respect to the plane of the planet’s orbit.

This means that the star will give rise to strong “obliquity tides” on the planet, and the dissipation of those tides could explain the decay of the orbit. For this to work something must be keeping the planet tilted over. Millholland suggests that a second planet in an outer orbit might be perturbing WASP-12b, keeping it in the high-obliquity state. This scenario requires some fine tuning, but if WASP-12 is the only system known to show this behaviour then the explanation is plausible.