by Ludwig Scheibe (TU Berlin), August 2024
Early Speculation
The possibility of other worlds out there has captivated humankind’s imagination for millennia, and there have always been individuals who were convinced that these worlds exist. Ancient Greek philosopher Epicurus wrote of “an infinite number of worlds, some like this world, some unlike it”, and medieval Persian scholar Fakhr al-Din al-Razi explains that an all-powerful god would have the ability to create “a million worlds beyond this world, such that each one of those worlds be bigger and more massive than this world”.
The Greek philosopher Epicurus developed, among many other philosophical concepts, a doctrine of nature according to which there are an infinite number of other worlds. Image source: Wikimedia Commons
Renaissance Italian Giardano Bruno even explains why we don’t see the other worlds: “There are countless Suns and countless Earths all rotating round their suns in exactly the same way as the seven planets of our system. We see only the suns because they are the largest bodies and are luminous, but their planets remain invisible to us because they are smaller and non-luminous.” German philosopher Immanuel Kant wrote in his Naturgeschichte: “All the fixed stars which the eye discovers in the hollow depths of the heavens, […] are suns and centers of similar systems.” However, these are predicated on plausibility arguments and none of these early pioneers could present direct evidence.
70 Ophiuchi – early claims of a possible exoplanet
Differences between observed motion and undisturbed orbitby 70 Ophiuchi. From: See, Astron. Journ. 16, 1896
One of the first claims of an exoplanet based on observational evidence was made in 1855 by William Jacob working at Madras observatory in India. He argued that the observed orbital motion of the binary star system 70 Ophiuchi could not be explained with just the two stars and calls it “highly probable” that there exists a “planetary body” in that system. This was expanded upon in 1899 by Thomas See at the University of Chicago. He calculated orbital characteristics for this companion, including an orbital period of 88 years and and eccentricity of 0.5. However, even in the same year, Forest Moulton showed that See’s orbit would not be stable and thus it was unlikely to be correct. Furthermore, several later observations of 70 Ophiuchi failed to reproduce Jacob’s and See’s observed perturbations.
Otto Struve – proposing radial velocity and transit method
Throughout the 20th century, there were several claims of planetary companions, most of them made in a similar fashion to Jacob and See: perturbations in a star’s position supposedly caused by an unseen planetary companion – what we would call astrometric detection today. None of these claims could convincingly be confirmed. However, in 1952 the astronomer Otto Struve, from a distinguished line of astronomers and then working at the University of California Berkeley, published a short paper that would prove to be highly influential and prescient.
Otto Struve, one of the fathers of modern exoplanet research with his proporals for promising detection methods. Image Source: NRAO/AUI/NSF
In the paper titled Proposal for a project of High-Precision Stellar Radial velocity work, Struve argued from planet formation standpoints that most stars should have planets. He also proceeded to propose both the radial velocity method for finding planets based on the “wobble” they induce in their host stars, measurable by the Doppler shift in the star’s spectrum, and the transit method, detecting a planet by the dip in brightness caused by the planet moving between us and the star. He also points out “It is not unreasonable that a planet might exist at a distance of 1/50 astronomical unit.”, which would make it just about detectable for state-of-the-art instruments at the time. It would take another 40 years, but Struve turned out to be absolutely correct both in the methods he favored and in his supposition that there would be very close-in planets – what we might call hot Jupiters and ultra-short-period planets today.
The tragedy of Bernard’s star
It would take a while for Struve’s recommendation and predictions to pan out. In the meantime, another astrometric detection took center stage and brought exoplanet’s back to public perception: In 1963, Peter van de Kamp of Spoul Observatory in the US published Astrometric Study of Barnard’s Star from Plates Taken with the 24-inch Spoul Refractor.
Peter van de Kamp was convinced – very probably mistakenly – that he had discovered at least one gas giant around Barnard’s Star. Image source: Wikimedia Commons
He presented 25 years of astrometric measurements of the star Gliese 699, better known as Barnard’s star – coincidentally also in the constellation of Ophiuchi, the same as Jacob’s and See’s binary system. The measurements show perturbations in the star’s position and van de Kamp concludes a companion of 1.6 times Jupiter’s mass and an orbital period of 24 years is responsible. Six years later, he published a refined analysis which now points to the existence of two planets of 1.1 and 0.8 Jupiter masses, on orbits of 26 and 12 years, respectively.
However, independent verification of the measurements was lacking. In fact, in 1973, George Gatewood and Heinrich Eichhorn published “An unsuccessful search for a planetary companion of Barnard’s star”, in which they present their own measurements of Bernard’s star and show that those do not reproduce van de Kamp’s results. Additionally, in the same year, John Hershey, who worked at the same observatory as van de Kamp published an analysis of the photographic plates that were the basis of van de Kamp’s result. Hershey concluded that what his colleague had seen was not, in fact, a perturbation in Bernard’s star’s motion, but an instrument error in the telescope. After further observation of the star with yet other telescopes also failed to reproduce the existence of the planet, it was widely accepted that the publicized discovery had in fact been in error.
Until his death in 1995, coincidentally the same year when the first extrasolar planet around a sun-like star was detected, Van de Kamp held on to his discovery and continued to publish papers refining the orbital characteristics. Much later, in 2018, a planet candidate around Barnard’s star was again published, but that has also since been disputed by several other teams. In any case, this candidate supposedly has a mass of only about 3 times that of Earth, with an orbital period of less than a year, which means it is definitely not the Jupiter-sized gas giant – or several gas giants – on multiple-year orbits that van de Kamp was claiming.
Tentative observations using the radial velocity
The story of Bernard’s star shows how important independent verification is in science, especially for extraordinary claims like the first planet around a star other than our Sun. Its somewhat disappointing conclusion may have served as a cautionary tale for subsequent claims, but the search continued. Now, radial velocity campaigns like the one advocated by Otto Struve became more common. In 1988, Campbell, Walker and Yang published a study of the radial velocity variations of 12 faint dwarf stars and found “For seven of 15 stars there is a ‘possible’ or ‘probable’ companion” and that these companions were “more closely related to planets than to brown dwarfs.” Note the cautious language here, the scientists clearly felt that their data did not yet warrant a definitive claim. In a similar vein, a team of scientists around David Latham announced an “unseen companion” around HD 114762, but could not confirm that it was definitely – or even probably – a planet.
PSR1257+12 and 51 Pegasi – genuine planetary systems
It is sometimes said that the first extrasolar planet was discovered in 1995. That is not quite true. In 1992, Aleksander Wolszczan and Dale Frail published their discovery of two planet-sized bodies around the pulsar star PSR1257+12. However, this star is a very old pulsating neutron star, and it was thought that the planet’s might have formed as a result of violent mass ejections in the star’s evolution. The star was named “Lich”, and the three planets – another one was discovered in 1994 – “Draugr”, “Poltergeist” and “Phobetor” in 2015 by the International Astronomical Union, owing to the pulsar’s status as a “dead” star.
In 2019, NASA presented the pulsar PSR B1257+12 and its three planets in the style of a classic horror film poster. The depiction of the planets as ‘undead worlds’ is based on the fact that they orbit the remaining core of a star that has already ‘died’. Image source: NASA-JPL/Caltech
The much more famous and impactful event happened in 1995: Using the radial velocity method, Didier Queloz and Michel Mayor found a planet around the star 51 Pegasi, which has almost the same physical characteristics as our sun.
Didier Queloz and Michel Mayor, the two discoverers of the first confirmed exoplanet around a sun-like star, 51 Pegasi b. Credit: L. Weinstein/Ciel et Espace Photos
The planet, however, is about half as heavy of Jupiter – still significantly heavier than Saturn and as such still very much a gas giant – on a 4-day orbit around its star, with an orbital distance of only 5% of that of Earth. This discovery, verified by two other teams within weeks of its announcement, was a big deal: This was not an old star, where the companions could be explained as a result of late-stage stellar evolution. This was a Sun-like star, in the prime of its main-sequence evolution, so the discovered companion was a genuine planet formed during the star’s genesis. But this strange planetary system was completely unlike anything known from the Solar System: its large mass and short orbit made it the first so-called hot Jupiter, a planetary class utterly unknown at the time. It turned out that Otto Struve had been completely correct: there were giant planets on very small orbits and they were found using the radial velocity method.
This discovery kicked off an age of exoplanet discoveries that is going and expanding to this day.
