≡ Menu

Proxima Centauri: Microlensing Yields New Data

It’s not easy teasing out information about a tiny red dwarf star, even when it’s the closest star to the Sun. Robert Thorburn Ayton Innes (1861-1933), a Scottish astronomer, found Proxima using a blink comparator in 1915, noting a proper motion similar to Alpha Centauri (4.87” per year), with Proxima about two degrees away from the binary. Finding out whether the new star was actually closer than Centauri A and B involved a competition with a man with a similarly august name, Joan George Erardus Gijsbertus Voûte, a Dutch astronomer working in South Africa. Voûte’s parallax figures were more accurate, but Innes didn’t wait for debate, and proclaimed the star’s proximity, naming it Proxima Centaurus.

The back and forth over parallax and the subsequent careers of both Innes and Voûte make for interesting reading. I wrote both astronomers up back in 2013 in Finding Proxima Centauri, but I’ll send you to my source for that article, Ian Glass (South African Astronomical Observatory), who published the details in the magazine African Skies (Vol. 11 (2007), p. 39). You can find the abstract here.

Image: Shining brightly in this Hubble image is our closest stellar neighbour: Proxima Centauri. Although it looks bright through the eye of Hubble, as you might expect from the nearest star to the Solar System, the star is not visible to the naked eye. Its average luminosity is very low, and it is quite small compared to other stars, at only about an eighth of the mass of the Sun. However, on occasion, its brightness increases. Proxima is what is known as a “flare star”, meaning that convection processes within the star’s body make it prone to random and dramatic changes in brightness. The convection processes not only trigger brilliant bursts of starlight but, combined with other factors, mean that Proxima Centauri is in for a very long life. Astronomers predict that this star will remain on the main sequence for another four trillion years, some 300 times the age of the current Universe. These observations were taken using Hubble’s Wide Field and Planetary Camera 2 (WFPC2). Credit: NASA/ESA.

It’s a long way from blink comparators to radial velocity measurements, the latter of which enabled our first exoplanet discoveries back in the 1990s, measuring how the gravitational pull of an orbiting planet could pull its parent star away from us, then towards us on the other side of the orbit, with all the uncertainties that implies. We’re still drilling into the details of Proxima Centauri, and radial velocity occupies us again today. The method depends on the mass of the star, for if we know that, we can then make inferences about the mass of the planets we find around it.

Thus the discovery of Proxima Centauri’s habitable zone planet, Proxima b, a planet we’d like to know much more about given its enticing minimum mass of about 1.3 Earths and an orbital period of just over 11 days. Radial velocity methods at exquisite levels of precision rooted out Proxima b and continue to yield new discoveries.

We’re learning a lot about Alpha Centauri itself – the triple system of Proxima and the central binary Centauri A and B. Just a few years ago, Pierre Kervella and team were able to demonstrate what had previously been only a conjecture, that Proxima Centauri was indeed gravitationally bound to Centauri A and B. The work was done using high-precision radial velocity measurements from the HARPS spectrograph. But we still had uncertainty about the precise value of Proxima’s mass, which had in the past been extrapolated from its luminosity.

This mass-luminosity relation is useful when we have nowhere else to turn, but as a paper from Alice Zurlo (Universidad Diego Portales, Chile) explains, there are significant uncertainties in these values, which point to higher error bars the smaller the star in question. As we learn more about not just other planets but warm dust belts around Proxima Centauri, we need a better read on the star’s mass, and this leads to the intriguing use to which Zurlo and team have put gravitational microlensing.

Here we’re in new terrain. The gravitational deflection of starlight is well demonstrated, but to use it, we need to have a background object move close enough to Proxima Centauri so that the latter can deflect its light. A measurement of this kind was recently made on the star Stein 3051 B, a white dwarf, using data from the Hubble instrument, the first use of gravitational lensing to measure the mass of a star beyond our Solar System. Zurlo and team have taken advantage of microlensing events at Proxima involving two background stars, one in 2014 (source 1), the other two years later (source 2), but the primary focus of their work is with the second event.

Using the Spectro-Polarimetric High-contrast Exoplanet REsearch instrument (SPHERE) at the Very Large Telescope at Cerro Paranal in Chile, the researchers observed Proxima Centauri and the background stars from March of 2015 to June of 2017. You can see Proxima in the image below, with the two background stars. In the caption, IRDIS refers to the near-infrared imager and spectrograph which is a part of the SPHERE/VLT installation.

Image: This is Figure 1 from the paper. Caption: IRDIS FoV for the April 2016 epoch. The image is derotated, median combined, and cleaned with a spatial filter. At the center of the image, inside the inner working angle (IWA), the speckle pattern dominates, in the outer part of the image our reduction method prevents the elongation of the stars’ point spread functions (PSFs). The bars in the lower right provide the spatial scale. North is up and East is to the left. Credit: Zurlo et al.

The extraordinary precision of measurement needed here is obvious, and the mechanics of making it happen are described in painstaking detail in the paper. The authors note that the SPHERE observations will not be further refined because the background star they call Source 2 is no longer visible on the instrument’s detector. Nonetheless:

The precision of the astrometric position of this source is the highest ever reached with SPHERE, thanks to the exquisite quality of the data, and the calibration of the detector parameters with the large population of background stars in the FoV. Over the next few years, Proxima Cen will be followed up to provide a better estimation of its movement on the sky. These data will be coupled with observations from HST and Gaia to take advantage of future microlensing events.

The results of the two-year monitoring program show deflection of the background sources’ light consistent with our tightest yet constraints on the mass of Proxima Centauri. The value is 0.150 solar masses, with possible error in the range of +0.062 to -0.051, or roughly 40%. This is, the authors note, “the first and the only currently possible measurement of the gravitational mass of Proxima Centauri.”

The previous value drawn from mass-luminosity figures was 0.12 ± 0.02 M. What next? While Source 2 may be out of the picture using the SPHERE installation, the authors add that Gaia measurements of the proper motion and parallax of that star may further refine the analysis. Future microlensing will have to wait, for no star as bright as Source 2 will pass within appropriate range of Proxima for another 20 years.

The paper is Zurlo et al., “The gravitational mass of Proxima Centauri measured with SPHERE from a microlensing event,” Monthly Notices of the Royal Astronomical Society Vol. 480, Issue 1 (October, 2018), 236-244 (full text). The paper on Proxima Centauri’s orbit in the Alpha Centauri system is Kervella et al., “Proxima’s orbit around α Centauri,” Astronomy & Astrophysics Volume 598 (February 2017) L7 (abstract).


Comments on this entry are closed.

  • Wojciech Jacyk May 24, 2022, 7:15

    Good news on hunt for Earth-like stars from China.
    There is now a second proposal for an exoplanet search telescope in China.

    “The Closeby Habitable Exoplanet Survey (CHES) mission, which Ji’s team has been working on for almost a decade, aims to monitor about 100 sun-like stars within a distance of 32 light years from the solar system, and measure tiny changes in their relative position in the sky to search for Earth-like planets around them(…)The CHES project expected to detect roughly 50 Earth-like planets or so-called super-Earths, he said.”

    This seems to be different proposal from Earth 2.0 telescope which scientists from Shanghai put forward:

    “China could soon begin its first space-based hunt for exoplanets, if a proposal from the Shanghai Astronomical Observatory (SAO) gets the go-ahead this summer.
    The Earth 2.0 Telescope would spend four years orbiting sun-Earth Lagrange point 2, about 930,000 miles (1.5 million kilometers) from Earth. There, it would fixed its seven telescopes on a portion of the sky toward the galactic center and watch for signs of dimming as planets transit, or pass in front of, a star as they orbit.
    The main targets are roughly Earth-size exoplanets with similar orbits around sun-like stars.” “The mission would follow up on observations of an area of space that NASA’s Kepler space telescope studied for nine years, but the Earth 2.0 Telescope would have a much greater field of view, meaning it would be able to observe a larger area and more stars, Ge said.”

    If even one of these proposals gets approval it will be a welcomed addition to the exoplanet search efforts.

  • Michael Lorrey May 24, 2022, 10:56

    Would be nice to see what, if any, the 25% greater mass of Proxima means for the habitable zone dimensions, and how that might affect the habitability of Prox b.

    • andy May 24, 2022, 12:49

      It’s worth noting the error bars here: the previous estimate is within the 1σ error bars on the microlensing estimate. Given the uncertainties, saying that the mass is greater by 25% is rather premature here.

  • Adam Crowl May 24, 2022, 16:37

    Hi Paul
    Previous mass estimates were from mass-luminosity relations derived from eclipsing binary stars, so this is a welcome direct confirmation of those figures, given that 0.12 solar masses is well within the error bars of the micro-lensing result. It’ll be interesting to see what the new metallicity fraction of the Sun does to such figures. The Solar metallicity crisis might be resolved, with new spectroscopic results for the Sun giving the Z fraction as 0.0225. That now matches the Helioseismology results and will have flow on effects to all stellar modeling.

  • Michael Christopher Fidler May 24, 2022, 22:19

    Great, we are finding ways to get more information out of a little pinhole of light! So if this holds out for the .150 sun mass would that make Proxima a M5 instead of a M5.5 or even a M4? What about doing this with all the other nearby red dwarfs and brown dwarfs that have background stars passing near them?

    MAY 24, 2022
    AI reveals unsuspected math underlying search for exoplanets.

    “In a paper appearing this week in the journal Nature Astronomy, the researchers describe how an AI algorithm developed to more quickly detect exoplanets when such planetary systems pass in front of a background star and briefly brighten it—a process called gravitational microlensing—revealed that the decades-old theories now used to explain these observations are woefully incomplete.”


    A Mathematical Treatment of the Offset Microlensing Degeneracy.

  • wdk May 24, 2022, 22:51

    When it was noted that Proxima Centauri’s separation from A and B is about 2 degrees of celestial arc, my mental picture of the place changed considerably. Often binaries have considerable separation but remain spectroscopic to observers. And then A-B are about out planet scales of separation which means a few dozen arc seconds separation.

    It is all consistent, of course, but when you look at a field of view which contains A&B as well as Proxima, the view encompasses a background of many stars. Discovering a connection between A-B and Proxima seems like a feat akin to Clyde Tombaugh’s discovery of Pluto. Many field stars that were background surrounding and in between.

    Call it “most likely mass” for lack of other designation. But if microlensing has caused this base to move from the mass luminosity estimate ( based on distance and parallax), then periods on planets will represent larger semi-major axes. Doppler estimates of masses depend on inclinations. Luminosity remains the same, but the perturbing planets move farther from the star. I.e., become cooler.

    Before I had accepted the notion that it was ten thousand or so AUs from the other two objects, but hadn’t grasped some of the consequences. It sent me to the Wikipedia page to brush up.