≡ Menu

A Laser ‘Comb’ for Exoplanet Work

It’s been years since I’ve written about laser frequency comb (LFC) technology, and recent work out of the Max Planck Institute of Quantum Optics, the Kiepenheuer Institute for Solar Physics and the University Observatory Munich tells me it’s time to revisit the topic. At stake here are ways to fine-tune the spectral analysis of starlight to an unprecedented degree, obviously a significant issue when you’re dealing with radial velocity readings of stars that are as tiny as those we use to find exoplanets.

Remember what’s happening in radial velocity work. A star moves slightly when it is orbited by a planet, a tiny change in speed that can be traced by studying the Doppler shift of the incoming starlight. That light appears blue-shifted as the star moves, however slightly, towards us, while shifting to the red as it moves away. The calibration techniques announced in the team’s paper show us that it’s possible to measure a change of speed of roughly 3 cm/s with their methods, whereas with conventional calibration techniques, the best measurement is roughly 1 m/s (although see the citations below for HARPS calibration of an LFC that reaches 2.5 cm/s). Detecting an Earth-mass planet in an Earth-like orbit around a solar-type star involves observing velocity changes of 10 cm/s or less, so we’re clearly entering the right range here.

Let’s back up and consider how a laser frequency comb works. Below is an image from the European Southern Observatory explaining the ‘comb’ analogy — as you can see, the graph resembles a fine-toothed comb, one built around short, equally spaced pulses of light created by a laser. The different colors of the pulsed laser light are separated based on their individual frequencies. Combining an ultrafast laser as a calibration tool with an external source of light allows scientists to measure the frequency of the external light to a high degree of precision.

ann12037a

Image: This picture illustrates part of a spectrum of a star obtained using the HARPS instrument on the ESO 3.6-metre telescope at the La Silla Observatory in Chile. The lines are the light from the star spread out in great detail into its component colours. The dark gaps in the lines are absorption features from different elements in the star. The regularly spaced bright spots just above the lines are the spectrum of the laser frequency comb that is used for comparison. The very stable nature and regular spacing of the frequency comb make it an ideal comparison, allowing the detection of minute shifts in the star’s spectrum that are induced by the motion of orbiting planets. Note that in this image, the colour range is for illustrative purposes only, as the real changes are much more subtle. Credit: ESO.

The laser frequency comb is, then, a standard ‘ruler’ that can measure the frequency of light to extreme precision. In the case of the recently announced findings, the researchers worked with sunlight averaged over the complete solar disk, as captured by the ChroTel solar telescope (located at the Vacuum Tower Telescope installation in Tenerife, Canary Islands). They combined this light with the light from the laser frequency comb, injecting both into a single optical fiber. The result was sent on to a spectrograph for analysis, with striking results. Lead author Rafael Probst (Max Planck Institute of Quantum Optics) comments:

“Our results show that if the LFC light and the sunlight are simultaneously fed through the same single-mode fibre, the obtained calibration precision improves by about a factor of 100 over a temporally separated fibre transmission. We then obtain a calibration precision that keeps up with the best calibration precision ever obtained on an astrophysical spectrograph, and we even see considerable potential for further improvement.”

Probst goes on to say that although the technique is currently restricted to solar spectroscopy, it should be workable even for faint astronomical targets as it is perfected. He comments in this news release from the Institute of Physics that a key aspect of the work is the clean and stable beam at the output that results from using single-mode fiber, a kind of fiber common in laser applications but relatively little used thus far in astronomy. The LFC at the Vacuum Tower Telescope is the first installation for astronomical use based on single-mode fiber.

These refinements of laser frequency comb technique point toward future measurements of Doppler shifts that will make detecting Earth-sized planets with radial velocity methods more likely. The laser frequency comb seems poised to become a major tool. “In astronomy, frequency combs are still a novelty and non-standard equipment at observatories,” the authors write in their conclusion. “This however, is about to change, and LFC-assisted spectroscopy is envisioned to have a flourishing future in astronomy.”

The paper is Probst et al., “Comb-calibrated solar spectroscopy through a multiplexed single-mode fiber channel,” New Journal of Physics Vol. 17 (February 2015) 023048 (abstract). See also this video abstract of the work. Laser frequency comb work at HARPS reaching into the cm/s range is reported in Wilken et al., “A spectrograph for exoplanet observations calibrated at the centimetre-per-second level,” Nature Vol. 485, Issue 7400 (May, 2012), 611-614 (abstract).

tzf_img_post

Comments on this entry are closed.

  • Andrew LePage February 25, 2015, 11:15

    While the laser comb and other improvements in hardware and observing techniques certainly promise to improve the accuracy of radial velocity measurements, one has to be VERY careful about the claims being made that this will allow the detection of of Earth-size extrasolar planets. The problem astronomers are beginning to encounter is that their ability to detect smaller planets is not being limited by their instruments but by the natural noise or “jitter” created by the star being observed as well as the ability of astronomers to accurately model and remove the effects of that noise from their data. It is the natural stellar “noise” caused by changes in magnetic surface activity that resulted in the discovery last year that most of the habitable planets thought to be orbiting GJ 581 and GJ 667C do not in fact exist and it was “jitter” masquerading as planets.

    http://www.drewexmachina.com/2014/07/07/the-disappearing-habitable-planets-of-gj-581/

    http://www.drewexmachina.com/2014/09/07/habitable-planet-reality-check-gj-667c/

    The limitations of the ability of astronomers to model stellar “jitter” is also at the heart of the controversy about the discovery of Alpha Centauri Bb which remains unconfirmed after over two years.

    http://www.drewexmachina.com/2014/08/11/the-search-for-planets-around-alpha-centauri/

  • Joy February 25, 2015, 16:08

    What Andrew said: +++++

  • Mike February 26, 2015, 14:16

    Of course nothing is easy in Astronomy. Noisier stars will require more observations. At the very least finer detailed spectroscopy will provide a better understanding of Asteroseismology.

    Since it’s the lower mass M-Dwarfs that are usually the noisier stars if they are hosting HZ Earth mass planets the radial velocity signal will be larger and perhaps still detectable despite the noise though likely requiring many observations.

    I think laser frequency comb technology is very promising.

  • Ron S February 26, 2015, 14:54

    Andrew, everything you say is correct yet the technique still would allow detection. It will just be difficult.

    If the “noise” is periodic or episodic it is possible to disentangle it from the periodic signals of planets. If the noise is long-term statistical (jitter) it is still possible to extract a periodic signal. In the latter case it could take an unreasonably long observation and would require integration techniques. There are constraints on what can be detected and what it will “cost” rather than a binary decision on whether or not this can work.

    If a low-sigma detection is made it can hopefully be combined with other detection methods (e.g. spectrographic absorption lines) to increase confidence.

    Every bit of data helps. Even if only to exclude ranges of exoplanet presence by mass and distance from the star.