A Super-Earth With an Air About It: 55 Cancri e, Janssen

NASA, ESA, CSA, Ralf Crawford (STScI) image: an artist's concept; the exoplanet 55 Cancri e and its sun, based on observations from NASA's James Webb Space Telescope and other observatories. (2024?)
Ralf Crawford’s impression of the exoplanet 55 Cancri e and its sun.

Screenshot from NASA's Eyes on Exoplanets: 55 Cancri e, and known planets in the Copernicus system. (2019?) (screenshot taken May 12, 2024)This month’s analysis of a piping hot Super-Earth’s atmosphere is a big deal.

But it’s not the “first” detection of a terrestrial exoplanet’s atmosphere, not by about eight years.1

I’ll be talking about how scientists sift through data, 55 Cancri e’s atmosphere, its planetary system, why 55 Cancri e — the exoplanet was officially named Janssen in 2015 — and why calling Janssen a “diamond planet” may be appropriate.


Scientists and 55 Cancri e: How They Know What They Know

NASA, ESA, CSA, Joseph Olmsted (STScI) Science: Aaron Bello-Arufe (JPL)'s illustration: 'A light curve of 7.5- to 11.8-micron light captured by NASA's James Webb Space Telescope's MIRI (Mid-Infrared Instrument) in March 2023 shows the decrease in brightness of the 55 Cancri system as the rocky planet 55 Cancri e moves behind the star, a phenomenon known as a secondary eclipse. ... indicates that heat is being distributed from the dayside to the nightside of the planet, possibly by a volatile-rich atmosphere.' (2024)
Illustration: 55 Cancri e’s secondary eclipse light curve, 7.5- to 11.8-microns. (March 2023, published 2024) NASA/ESA/CSA/STScI/JPL.

The Copernicus (55 Cancri A) planetary system is just like the Solar System. Except for how it’s different.

NASA’s Webb Hints at Possible Atmosphere Surrounding Rocky Exoplanet
NASA Webb Mission Team, Goddard Space Flight Center, NASA (May 8, 2024)

“Researchers using NASA’s James Webb Space Telescope may have detected atmospheric gases surrounding 55 Cancri e, a hot rocky exoplanet 41 light-years from Earth. This is the best evidence to date for the existence of any rocky planet atmosphere outside our solar system.

“Renyu Hu from NASA’s Jet Propulsion Laboratory (JPL) in Pasadena, California, is lead author on a paper published today in Nature. ‘Webb is pushing the frontiers of exoplanet characterization to rocky planets,’ Hu said. ‘It is truly enabling a new type of science.’…”

I’m not sure exactly what “new type of science” Dr. Hu had in mind. This Dr. Hu is the one with a PhD in planetary science from MIT, by the way; not the nuclear physicist who retired in 1994.

Screenshot and link to 'A secondary atmosphere on the rocky exoplanet 55 Cancri e';  Renyu Hu, Aaron Bello-Arufe, Michael Zhang, Kimberly Paragas, Mantas Zilinskas, Christiaan van Buchem, Michael Bess, Jayshil Patel, Yuichi Ito, Mario Damiano, Markus Scheucher, Apurva V. Oza, Heather A. Knutson, Yamila Miguel, Diana Dragomir, Alexis Brandeker, Brice-Olivier Demory; submitted May 8, 2024; arXiv, Cornell UniversityAt any rate, this Dr. Hu is the lead author of “A secondary atmosphere on the rocky Exoplanet 55 Cancri e”, published in Nature on May 8, 2024.

And, although Nature is a peer-reviewed scientific journal that’s currently unavailable to me, I do have access to a very recent pre-press draft on Cornell’s open-access arXiv service.1 I’ve talked about paywalls, member-only online resources, and other frustrations, before.

Getting back to “a new type of science”, the data Dr. Hu’s team was working with is new: partly from 2023 observations by the James Webb Space Telescope.

But as far as I could tell, they’re not using new mathematical tools. On the other hand, I could have missed something. I’m taking it easy this week, and didn’t pore over their pre-print paper all that intently.

Bayesian Basics and Dealing With Incomplete Data

Gnathan87's chart: results for an archaeological simulation, an example of Bayesian inference. (October 2011) via Wikimedia Commons, used w/o permissionMaybe part of the “new science” is running data through several analyses, with a different team working on each analysis.

“…The NIRCam data were analyzed by 4 independent teams with different pipelines (Methods). We removed 1/f noise (the correlated read noise that impacts data across a wide range of timescales with a 1/f power spectrum33)…”
(“A secondary atmosphere on the rocky exoplanet 55 Cancri e’“; Renyu Hu, Aaron Bello-Arufe, et al.; (submitted May 8, 2024) via arXiv, Cornell University)

Besides spreading out the workload, independent analysis teams should lower the odds that folks who are crunching the numbers will unintentionally bias their results. That can be a problem for any sort of analysis, not just science stuff, and that’s another topic.

One of the tools Hu and company used is Bayesian math: which I hadn’t noticed in research papers until fairly recently. But then, I’m not a scientist; and it wasn’t until fairly recently that I could get my virtual hands on such things.

Bayesian statistics is what happened when Pierre-Simon Laplace started working on Bayes’ theorem. Bayes was a statistician, philosopher, and Presbyterian minister. A whole bunch of folks have fine-tuned what we call Bayesian statistics, but I’ll skip all that.

Basically, Bayesian statistics lets scientists work out the odds that something is true, based on facts that may or may not matter. It’s useful when we don’t know everything.2

Which, arguably, is pretty much always the case.

Studying Starlight: Transits, Eclipses, and a Whole Lot of Math

Illustration from NASA/ESA/CSA/Joseph Olmsted (STScI), Science by Renyu Hu (JPL), Aaron Bello-Arufe (JPL), Michael Zhang (University of Chicago), Mantas Zilinskas (SRON Netherlands Institute for Space Research). 'A thermal emission spectrum of the super-Earth exoplanet 55 Cancri e, captured by NASA's James Webb Space Telescope's NIRCam (Near-Infrared Camera) GRISM Spectrometer (F444W) and MIRI (Mid-Infrared Instrument) Low-Resolution Spectrometer, shows that the planet may be surrounded by an atmosphere rich in carbon dioxide or carbon monoxide and other volatiles, not just vaporized rock.' (2024)
Illustration: 55 Cancri e’s thermal emission spectrum from NIRCam, GRISM Spectrometer (F444W), & MIRI. (March 2023, published 2024) NASA/ESA/CSA/STScI/JPL.

Studying 55 Cancri e would be fairly easy, if it was in the Solar System. We’d just point a telescope toward the planet and take a few pictures.

NASA/JPL-Caltech (R. Hurt/IPAC's infgraphic: 'scientists used the James Webb Space Telescope to observe the exoplanet WASP-18 b and its star before, during and after the planet was eclipsed. By measuring the change in light when the planet travels behind the star, the planet's brightness is revealed. From these measurements, scientists were able to make a temperature map of the planet's day side. Displayed temperature range: 2,800 to 4,800 degrees Fahrenheit.' May 31, 2023)If 55 Cancri e’s edges were fuzzy, that’d mean it’s got an atmosphere.

Scientists could put a spectrometer on the telescope, look at what wavelengths get reflected and/or absorbed, and that’d tell them what’s in the atmosphere or on the surface.

Or they could arrange for a probe to be dropped into 55 Cancri e’s atmosphere, and get data from that.

Just one problem. Light from 55 Cancri e takes 41 years to get here. Astronomers are doing well to work out which parts of the 55 Cancri A system’s light are coming from planets, and which are from the star.

Happily, since 55 Cancri e passes in front of and behind its star during each orbit, careful observations tell scientists quite a bit about the planet.

The trick is measuring light when:

  • Both are visible and neither is blocked
  • The planet blocks part of the star’s light
  • The star blocks light reflected by the planet

Then, using a whole lot of math, scientists work out what’s reflected from the just planet. And what is (or isn’t) shining through the planet’s atmosphere: if it’s got one.

I talked about this last year, along with what we’re learning about weather — winds, specifically — on WASP-18b.3


Welcome to the Copernicus Planetary System

Screenshot from NASA's Eyes on Exoplanets: 55 Cancri e, and known planets in the Copernicus system. (2019?) (screenshot taken May 12, 2024)
The star Copernicus, 55 Cancri A, and its known planets.

Star chart by Roger Sinnott, Rick Fienberg (IAU/Sky and Telescope magazine): the constellation Cancer.55 Cancri is a double star, between Rho2 and Iota Cancri in our sky.

55 Cancri B is a red dwarf with no name.

But in 2015, the IAU made it official: 55 Cancri A’s name is Copernicus.

Copernicus / 55 Cancri A’s planets are:

  • Galileo (b)
  • Brahe (c)
  • Lipperhey (d)
  • Janssen (e)
  • Harriot (f)

55 Cancri has a Bonner Durchmusterung designation, BD+28°1660, which strongly suggests that it was known at least as far back as the mid-19th century.

Scientists started spotting planets around 55 Cancri A in the late 20th century.

Galileo, 55 Cancri b, was discovered in 1996; Janssen, 55 Cancri e — the planet I’m talking about this week — was discovered in 2004.

Harriot’s discovery, that’s 55 Cancri f, was announced in 2005 and published in 2007.

At that point, the Copernicus planetary system had five known planets. The last I checked, there’s informed speculation that there may be more.

So how come one of last week’s headlines announced the “discovery” of Janssen??

I suspect deadline pressures are a factor, along with the need to grab attention.

That may account for headlines like these:

That’d be impressive, if other scientists hadn’t published this research, back in 2016:

  • “Detection of an atmosphere around the super-Earth 55 Cancri e”
    A. Tsiaras et al., The Astrophysical Journal (March 24, 2016)

From 'The Fifth Element', via IMDB.com, used w/o permission.In my darker moments, I feel that many news editors got their science education by watching “Captain Planet and the Planeteers” and “The Fifth Element”.4

More likely, the headlines and articles reflect each publication’s readership: 0r editorial perceptions thereof.

The “Diamond-covered” description isn’t entirely inaccurate. I’ll get back to that.

There’s No Place Like Home: But the Copernicus System Comes Close

Screenshot from NASA's Eyes on Exoplanets: 55 Cancri e and inner planets of the Copernicus system. (2019?) (screenshot taken May 12, 2024)
A closer look at the Copernicus planetary system.

NASA's diagram, comparing Cancri 55 planetary system and the Solar System's Earth and Jupiter. (2006)Although the Copernicus planetary system has a Jupiter-sized planet orbiting at about Jupiter’s distance, it’s not quite like our Solar System.

It is, however, the second-closest match we’ve found, as far as I know.

Second Planetary System Like Ours Discovered
Shannon Hall, Universe Today (November 27, 2013)

“…KOI-351 is ‘the first system with a significant number of planets (not just two or three, where random fluctuations can play a role) that shows a clear hierarchy like the solar system — with small, probably rocky, planets in the interior and gas giants in the (exterior),’….”
[emphasis mine]

Our Solar System’s Cousin?
NASA/JPL-Caltech (November 6, 2007)

“…The 55 Cancri system is currently the closest known analogue to our solar system, yet there are some fundamental differences.

“The similarities begin with the stars themselves, which are about the same mass and age. Both stars also host big families of planets….

“…In addition, both planetary systems have giant planets in their outer regions. The giant located far away from 55 Cancri is four times the mass of our Jupiter, and completes one orbit every 14 years at a distance of five times that between Earth and the sun … Our Jupiter completes one orbit around the sun every 11.9 years, also at about five times the Earth-sun distance….”
[emphasis mine]

Other stars, like HD 70642 and HIP 11915, have roughly Jupiter-mass planets orbiting about as far out as Jupiter. But again: the Copernicus system is still among the very few that resemble our Solar System.

Make that vaguely resemble.

The KOI-351 system — it’s also called Kepler-90, has a mess of other designations, and if I start talking about that, this won’t be ready by Saturday.

Anyway, the KOI-351/Kepler-90 system has eight planets.

The smaller ones orbit close to their star, which is almost but not quite like ours. So far, the KOI-351 system sounds just like the Solar System. Except that all eight planets are closer to their sun than Earth is to ours.

The Copernicus / 55 Cancri planetary system has a roughly Jupiter-mass planet in an orbit roughly as big as Jupiter’s.

But the other known planets aren’t arranged like the Solar System’s:with smaller, rocky, worlds close to the star and giant planets farther out.

Here’s the known Copernicus system planets, starting with the innermost one:5

  • e (Janssen) — 7.99 times Earth’s mass, 1.875 times Earth’s diameter
  • b (Galileo) — 0.8 times Jupiter’s mass, maybe more
  • c (Brahe) — 51.2 times Earth’s mass, maybe more
  • f (Harriot) — 49.8 times Earth’s mass, maybe more
  • d (Lipperhey) — 3.12 times Jupiter’s mass, maybe more

I left out symbols like ±, M[astronomical symbol meaning “Earth”], and MJ, which say which Solar System planet is being used for comparison, and how accurate our data is. Basically, those numbers are approximations, but pretty close.

Copernicus: Giant Planets and a Super-Earth Circling a Slightly Strange Star

Chaos syndrome's illustration, comparing orbits of 55 Cancri A's planetary sysytem and the Inner Solar System's.Lipperhey, the outermost known planet in the Copernicus system, is roughly three and an eighth times Jupiter’s mass.

The inner planets — Janssen, Galileo, Brahe, and Harriot — all have orbits smaller than Earth’s.

Janssen, the one I’m talking about today, whips around Copernicus once every 17 hours and 41 minutes. Just under 17 hours and 41 minutes, actually. The point is that it’s really, really, close to its sun.

On top of that, Copernicus is a slightly odd star. Although is it’s a trifle cooler and less massive than our sun, Copernicus apparently puts out a bit more energy than a K0-V main sequence star should.

So it’s classed as K0IV-V: maybe on the main sequence, maybe a subgiant star.

More oddities: Copernicus has more “metals” than our sun. In astronomer-speak, a “metal” is any element heavier than hydrogen or helium. Copernicus has 186% the solar amount of iron; and a carbon/oxygen ratio of 0.78, compared to our star’s 0.55.

All that apparently makes the age of Copernicus hard to work out. But, whether it’s 7,400,000,000 years old or 12,700,000,000 years old, it’s been around considerably longer than the Solar System.

Janssen isn’t quite so ambiguous. It’s a super-Earth that actually is a terrestrial planet: a rocky (?) world, like Earth. It’s twice our home’s diameter, and so hot that it might have had an atmosphere of vaporized rock.6


Janssen: ‘Terrestrial’, But Not Like Earth

Renyu Hu et al.: figure 7 'Thermal emission spectra of 55 cnc e if it has a thin, vaporized-rock atmosphere. a. The spectrum is calculated with the model of Ref32 for varied silicate-based melt compositions. b. The spectrum is calculated with the model of Ref31 assuming the BSE composition as magma composition. The NIRCam spectrum is shown with a mean eclipse depth of 150 ppm in both panels. The difference between the two models is mainly due to the different opacities used for SiO, but regardless a vaporized-rock atmosphere is inconsistent with the MIRI measured spectrum.' (2024) via arXiv, used w/o permission
Comparing Janssen’s hypothetical rock vapor atmosphere with NIRCam spectrum. Renyu Hu et al. (2024)

Again, Janssen is almost certainly a terrestrial planet: like Mercury, Venus, Earth-Moon, and Mars in the Solar System. I’m inclined to see the Earth-Moon system as a double planet; which is yet another topic.

But Janssen’s ‘rocks’ may not be the silicate sort we’re familiar with.

Now, about Janssen’s atmosphere. Thanks to this month’s “A secondary atmosphere on the rocky exoplanet 55 Cancri e” paper, we can be pretty sure Janssen has an atmosphere; and that Janssen’s air isn’t made of rock vapor.

Odds are that Janssen’s atmosphere has a fair amount of carbon dioxide or carbon monoxide, something that’s mentioned in the study’s opening Summary Paragraph:

“…The measurements rule out the scenario where the planet is a lava world shrouded by a tenuous atmosphere made of vaporized rock29-32, and indicate a bona fide volatile atmosphere likely rich in CO2 or CO. This atmosphere can be outgassed from and sustained by a magma ocean….”
(“A secondary atmosphere on the rocky Exoplanet 55 Cancri e“, Renyu Hu et al., preprint (May 2024) via arXiv)

Black body radiation curve, Astronomy Education at the University of Nebraska-Lincoln.They used data from the James Webb Space Telescope’s MIRI (Mid-Infrared Instrument) and NIRCam (Near-InfraRed Camera) — I talked about JWST and infrared astronomy last December.

I remember reading about carbon dioxide and monoxide in Janssen’s atmosphere in science news articles: but not an odd chemical mentioned in the research paper’s main text:

“…The presence of H2O, SO2, or PH3 could improve the fit to the spectral modulation in 4-5 μm in some cases. In the other set of models, we assumed an atmosphere in volatile equilibrium with the underlying magma ocean44,45….

“…In summary, the best-fit models center around three possible scenarios: a CO2-rich atmosphere (C+O~10-3, C/O<1), a CO-dominated atmosphere (C+O~1, C/O>=1), or a PH3-rich atmosphere with minimal C+O influence (C/O=1, C+O=10-7, PH3~10-4). The first scenario is uniquely favored when fitting the NIRCam data alone (Extended Data Fig. 8), which is consistent with spectral retrievals. By contrast, the MIRI data does not indicate any clear molecular features, suggesting either efficient heat redistribution or overlapping absorption features (e.g., H2O in 7-9 μm and CO2 in 9-11 μm) that place the photosphere to the cooler regions of the atmosphere….”
(“A secondary atmosphere on the rocky Exoplanet 55 Cancri e“, Renyu Hu et al., preprint (May 2024) via arXiv) [emphasis mine]

Phosphine, PH3, is a compound of phosphorus and hydrogen. It’s highly toxic, and used for both pest control and microelectronics manufacturing.

Other than what Hu et al. said about adding a dash of phosphine for a better fit, I don’t see why they picked that particular compound.

Hydrogen is by far the most common element in this universe. Carbon and oxygen are both among the 10 most common elements, at least in this galaxy. Phosphorus isn’t.

My guess is that someone will crunch numbers for the “PH3-rich atmosphere with minimal C+O influence” atmosphere model.

I suspect it’s just a matter of time before a reporter notices PH3, phosphine, in one of the study’s models; and remembers the occasional published reports of phosphine in the atmosphere of Venus.7

Although phosphine might be a biosignature, the odds of life on Venus are almost nil, and they’re even less for Jannsen.

Diamonds are another matter.

Like a Diamond in the Sky?

Haven Giguere's illustration: 'the interior of 55 Cancri e - an extremely hot planet with a surface of mostly graphite surrounding a thick layer of diamond, below which is a layer of silicon-based minerals and a molten iron core at the center.' (2012)
Haven Giguere’s illustration: 55 Cancri e as a ‘diamond planet’. YaleNews (2012)

Nearby super-Earth likely a diamond planet
“New research led by Yale University scientists suggests that a rocky planet twice Earth’s size orbiting a nearby star is a diamond planet.”
YaleNews (October 11, 2012)

“New research led by Yale University scientists suggests that a rocky planet twice Earth’s size orbiting a nearby star is a diamond planet.

“‘This is our first glimpse of a rocky world with a fundamentally different chemistry from Earth,’ said lead researcher Nikku Madhusudhan, a Yale postdoctoral researcher in physics and astronomy. ‘The surface of this planet is likely covered in graphite and diamond rather than water and granite.’…”

I said I’d get back to this:

Each time scientists publish something about 55 Cancri e, Janssen, I can count on headlines like these popping up:

As I said earlier: reporters and editors deal with deadlines, headlines are supposed to grab attention, and Janssen was discovered in 2004. Under the circumstance, I’m impressed when an article clarifies whether “big” refers to width or mass.

“…The width of the planet is about twice that of Earth and is around 9 times heavier than Earth. According to the information shared by NASA, this exoplanet is known as 55 Cancri e….”
(“NASA Discovers Exoplanet Made Of Diamond And 9 Times Bigger Than Earth“, Curated by Buzz Staff, News18, Delhi, India (May 11, 2024))

However, I’m not clear on where “the information shared by NASA” came from. NASA does have 55 Cancri e-related content, but it also gives Janssen’s mass as “7.99 Earths”.8

Maybe the “around 9 times heavier” thing came from truncating “7.99″. I don’t know.

Carbon Planets: Carbides and Maybe Diamonds

Gregg Dinderman's illustration in Sky and Telescope, comparing structure of a carbon planet and a silicate planet. Source: Marc J. Kuchner / Sara Seager. (2005)I have to admit that ‘diamond planet’ is a catchy phrase.

And Janssen may, in fact, have “a fundamentally different chemistry from Earth”. Which isn’t, actually, a new idea.

The Solar System’s inner planets are mostly silicates, “rocks”, and metals like iron. Much of the stuff we call rocks are silicates: compounds of oxygen and silicone.

At least since 2005, scientists have been saying that an exoplanet’s “rocks” might be made of elements other than oxygen and silicone. Like, for example, carbon and silicone. A “carbon planet” could have an iron-rich core with a mantle of silicon carbide.

And, if there’s enough pressure down where the planet’s mostly carbon, there could be diamonds instead of graphite.

Since Janssen’s sun has significantly more carbon than ours, it may really be a “diamond planet”.9 If so, finding proof will take time.

More, mostly about planets and stars:


1 Scientists, research, and a scientific journal:

  • Wikipedia
  • A secondary atmosphere on the rocky Exoplanet 55 Cancri e
    Renyu Hu, Aaron Bello-Arufe, Michael Zhang, Kimberly Paragas, Mantas Zilinskas, Christiaan van Buchem, Michael Bess, Jayshil Patel, Yuichi Ito, Mario Damiano, Markus Scheucher, Apurva V. Oza, Heather A. Knutson, Yamila Miguel, Diana Dragomir, Alexis Brandeker, Brice-Olivier Demory; preprint draft of paper published in Nature (May 8, 2024) (submitted May 8, 2024) via arXiv, Cornell University
  • First Detection of Super-Earth Atmosphere
    heic1603 — Science Release, Hubble Space Telescope News, ESA/Hubble (February 16, 2016)
  • Renyu Hu (Dr. Renyu Hu: Ph.D. in planetary science MIT (2013); M.S. Astrophysics, Tsinghua University (2009); Diplome d’Ingenieur, Ecole Centrale Paris (2009); B.S. Mathematics and Physics, Tsinghua University (2007))
  • Renyu Hu, PhD
    Renyu Hu’s Homepage

2 Statistics and minimizing errors:

3 Studying distant worlds:

4 Stars, planets, research (plus a cartoon and a movie); this was not hard to find:

5 Planets and planetary systems:

6 Planets, stars, and informed speculation:

7 Elements, compounds, and abiotic processes:

8 An exoplanet, the news, and science:

9 Science and informed speculation:

How interesting or useful was this post?

Click on a star to rate it!

Average rating 0 / 5. Vote count: 0

No votes so far! Be the first to rate this post.

I am sorry that this post was not useful for you!

Let me learn why!

How could I have made this more nearly worth your time?

About Brian H. Gill

I was born in 1951. I'm a husband, father and grandfather. One of the kids graduated from college in December, 2008, and is helping her husband run businesses and raise my granddaughter; another is a cartoonist and artist; #3 daughter is a writer; my son is developing a digital game with #3 and #1 daughters. I'm also a writer and artist.
This entry was posted in Science News and tagged , , , , , . Bookmark the permalink.

Thanks for taking time to comment!