One of the most fascinating topics in modern astronomy is the search for exoplanets, i.e. those planets that orbit a star other than the Sun. How is this research connected to laser technology? And to this paper in particular? This work is an interesting example on how two different disciplines, which only have in common the art of detecting and manipulating light, can assist each other.
Let's see in which way.

It is well known that direct observation of an exoplanet is almost impossible due to its extremely large distance from our solar system; however, its existence can be inferred and partially characterized by the indirect traces left by its movement. During its orbital motion, the planet changes the position and velocity of its parent star as they both orbit a common center of mass. The orbital motion usually causes radial velocity variations of the star with respect to Earth by several kilometers per second, and surprisingly it is possible to detect and evaluate the amplitude of such periodic movement also from very, very far away. This is possible thanks to the so-called Doppler effect, by which the measured frequency of a wave emitted by a source changes when the source moves relative to the observer; that's why, for example, we hear that the roar of a racing car changes pitch when the car rapidly approaches or recedes from us. The effect also takes place for the case of electromagnetic waves, such as light, where a change of pitch corresponds to a shift of its spectral components. Thanks to the Doppler effect, during the periodic motion of the star the frequency of its known emission lines decreases if the star recedes (redshift) and increases if it approaches (blueshift) the Earth.

Is it really possible to detect such spectral shifts and deduce the movement of stars that may be light-years away far from Earth? It is, but it requires the use of an astrophysical spectrograph with precise and stable wavelength calibration. This calls for a wavelength calibrator, which is a light source with high-density, regularly-spaced and stable optical lines over a broad bandwidth. And here comes the request from astrophysicists to the laser community to develop such a device. Since 1995, thanks to the help from high-resolution spectroscopy, nearly 2000 exoplanets were discovered, but this research requires to explore many different spectral regions, including the infrared. Here comes the device described in this paper: among other interesting applications, it also satisfies many requirements for a wavelength calibrator of astrophysical spectrographs.

The first devices for the generation of narrowband emission lines were based on lamps, but they suffered from thermal instabilities, which limited their spectral accuracy. The laser allowed to obtain spectral lines that had extremely high frequency definition and stability. With the use of broadband gain media and of the mode-locking technique, lasers have been able to emit sequences of pulses, spectrally corresponding to a large collection of equally spaced lines, also known as frequency combs. In this case, the spectral spacing is the repetition rate of the pulses, which inversely depends on the laser cavity length and hence on its optical resonances; on the other hand, the lines number and frequency is dictated by the gain band of the active medium. Combs with 1 GHz spacing can be generated by short laser cavities, but the limited tuning of the emission of their gain media makes them applicable as wavelength calibrators only over restricted spectral regions. How to tune the comb to a specific frequency region, addressing the calibration required by astrophysics? One interesting approach is to start from conventional mode-locked lasers, and frequency shift their emission; by frequency doubling in a suitable nonlinear crystal, for example, it is easy to tune the comb to the laser second harmonic, at high frequencies. How to obtain a comb at frequencies lower than the pumping laser? A very promising approach is the one employed in this paper, and is based on the Optical Parametric Oscillator (OPO). Inspired by lasers, OPOs are optical cavities equipped with a gain medium; however gain is not based on stimulated emission from atomic lines, but on parametric amplification. This is a nonlinear process where amplification occurs thanks to nonlinear interaction between a pump and two low-frequency beams, the signal and the idler If the cavity is optimized for high throughput at signal or idler frequencies, the amplification allows efficient energy flux from the pump pulses to the cavity modes at lower frequencies; this process is very flexible since it exhibits much larger emission tunability than traditional gain media. OPOs can be pumped by many types of lasers, including pulsed mode-locked lasers. Starting from their first development some years ago, many groups focused on the optimization of OPOs, leading to improvements in terms of repetition rate, tunability, emission bandwidth or stability of its modes.

Operation at GHz rate is particularly challenging because the peak power of the pump pulses emitted by mode-locked lasers decreases with increasing repetition rate, and makes it harder to trigger the nonlinear process required by parametric amplification; on the other hand, the average power may be very high, leading to surface damage of the gain medium. Balancing between high average power and low peak intensity calls for the search of suitable nonlinear crystals, with large enough nonlinear response and high damage threshold. The choice of the crystal is further limited by the amplification bandwidth, which should be broad enough to accommodate a large number of cavity modes. In the case of nonlinear processes, their bandwidth is dictated by the phase-matching condition, which depends on material dispersion and crystal thickness. The goal of the authors of this paper was to obtain broadband emission at the wavelength of 1.6 microns starting from a Ti:sapphire laser at 1 GHz repetition rate, therefore extending the tunability of 800-nm, high-repetition-ratecombs into the infrared. By exploring various amplification regimes, the authors were able to identify a medium, namely a 0.6-mm thick periodically-poled KTP, which has broadband amplification capabilities and large nonlinear figure of merit, particularly suited to compensate the low peak intensities of GHz operation. Broadband emission around the wavelength of 1.6 micron was achieved by operating the cavity at degeneracy, where signal and idler lay in the same spectral region. Degeneracy also has the important additional consequence that the cavity is resonant to both modes, leading to more efficient extraction of energy from the pump. By merging the advantages of degeneracy with the high repetition rate and the advantages of the crystal, the authors were able to obtain very promising preliminary results in terms of amplification bandwidth and pulse duration at GHz repetition rates. In the frequency domain, this corresponds to the emission of a broadband frequency comb at the wavelength of 1.6 micron, with 1-GHZ spacing. Each of these results separately was also obtained by other groups, but this work for the first time shows that, by a suitable design of the cavity, all these features can be simultaneously achieved. Further developments are still necessary in order to obtain a broadband absolute frequency reference, such as the stabilization and characterization of the carrier-envelope frequency offset, and further broadening of its emission. However, this device promises relevant improvements to direct comb spectroscopy and, in combination with sources with different repetition rates, to the generation of couple of pulses with variable delays for asynchronous optical sampling. In the future, it may also play an important role for the calibration of astronomical spectrographs in the infrared, contributing to the fascinating study of exoplanets.

---

---

 Add Comment

You must log in to add comments.