Metrology experiments can be limited by the noise produced by the laser involved via small fluctuations in the laser’s power or frequency. Typically, active power stabilization schemes consisting of an in-loop sensor and a feedback control loop are employed. Those schemes are fundamentally limited by shot noise coupling at the in-loop sensor. In this Letter, we propose to use the optical spring effect to passively stabilize the classical power fluctuations of a laser beam. In a proof of principle experiment, we show that the relative power noise of the laser is stabilized from approximately 2 × 10−5Hz−1/2to a minimum value of 1.6 × 10−7Hz−1/2, corresponding to the power noise reduction by a factor of 125. The bandwidth at which stabilization occurs ranges from 400 Hz to 100 kHz. The work reported in this Letter further paves the way for high power laser stability techniques which could be implemented in optomechanical experiments and in gravitational wave detectors.
This Letter reports the experimental realization of a novel, to the best of our knowledge, active power stabilization scheme in which laser power fluctuations are sensed via the radiation pressure driven motion they induce on a movable mirror. The mirror position and its fluctuations were determined by means of a weak auxiliary laser beam and a Michelson interferometer, which formed the in-loop sensor of the power stabilization feedback control system. This sensing technique exploits a nondemolition measurement, which can result in higher sensitivity for power fluctuations than direct, and hence destructive, detection. Here we used this new scheme in a proof-of-concept experiment to demonstrate power stabilization in the frequency range from 1 Hz to 10 kHz, limited at low frequencies by the thermal noise of the movable mirror at room temperature.
more » « less- Award ID(s):
- 1806634
- NSF-PAR ID:
- 10221609
- Publisher / Repository:
- Optical Society of America
- Date Published:
- Journal Name:
- Optics Letters
- Volume:
- 46
- Issue:
- 8
- ISSN:
- 0146-9592; OPLEDP
- Page Range / eLocation ID:
- Article No. 1946
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
more » « less
-
Abstract Using plasma mirror injection we demonstrate, both analytically and numerically, that a circularly polarized helical laser pulse can accelerate highly collimated dense bunches of electrons to several hundred MeV using currently available laser systems. The circular-polarized helical (Laguerre–Gaussian) beam has a unique field structure where the transverse fields have helix-like wave-fronts which tend to zero on-axis where, at focus, there are large on-axis longitudinal magnetic and electric fields. The acceleration of electrons by this type of laser pulse is analyzed as a function of radial mode number and it is shown that the radial mode number has a profound effect on electron acceleration close to the laser axis. Using three-dimensional particle-in-cell simulations a circular-polarized helical laser beam with power of 0.6 PW is shown to produce several dense attosecond bunches. The bunch nearest the peak of the laser envelope has an energy of 0.47 GeV with spread as narrow as 10%, a charge of 26 pC with duration of ∼ 400 as, and a very low divergence of 20 mrad. The confinement by longitudinal magnetic fields in the near-axis region allows the longitudinal electric fields to accelerate the electrons over a long period after the initial reflection. Both the longitudinal E and B fields are shown to be essential for electron acceleration in this scheme. This opens up new paths toward attosecond electron beams, or attosecond radiation, at many laser facilities around the world.more » « less
-
more » « less
State-of-the-art optical oscillators employing cryogenic reference cavities are limited in performance by the Brownian thermal noise associated with the mechanical dissipation of the mirror coatings. Recently, crystalline Al1−
x Gax As/GaAs coatings have emerged as a promising candidate for improved coating thermal noise. We present measurements of the frequency noise of two fully crystalline cryogenic reference cavities with Al0.92Ga0.08As/GaAs optical coatings. We report on birefringent noise associated with anticorrelated frequency fluctuations between the polarization modes of the crystalline coatings and identify variables that affect its magnitude. Comparing the birefringent noise between the two cryogenic reference cavities reveals a phenomenological set of scalings with intracavity power and mode area. We implement an interrogation scheme that cancels this noise by simultaneous probing of both polarization modes. The residual noise remaining after this cancellation is larger than both cavities’ thermal noise limits but still lower than the instabilities previously measured on equivalent resonators with dielectric coatings. Though the source of these noise mechanisms is unclear, we demonstrate that crystalline coatings can provide stability and sensitivity competitive with resonators employing dielectric coatings. -
more » « less
The Pound–Drever–Hall (PDH) cavity-locking scheme has found prevalent uses in precision optical interferometry and laser frequency stabilization. A form of frequency modulation spectroscopy, PDH enjoys superior signal-to-noise recovery, large acquisition dynamic range, wide servo bandwidth, and robust rejection of spurious effects. However, residual amplitude modulation at the signal frequency, while significantly suppressed, still presents an important concern for further advancing the state-of-the-art performances. Here we present a simplified and improved scheme for PDH using an acousto-optic modulator to generate digital phase reference sidebands instead of the traditionally used electro-optic modulator approach. We demonstrate four key advantages: (1) the carrier and two modulation tones are individually synthesized and easily reconfigured, (2) robust and orthogonal control of the modulated optical field is applied directly to the amplitude and phase quadratures, (3) modulation synthesis, demodulation, and feedback are implemented in a self-contained and easily reproducible electronic unit, and (4) superior active and passive control of residual amplitude modulation is achieved, especially when the carrier power is vanishingly low. These distinct merits stimulate new ideas on how we optimally enact PDH for a wide range of applications.
-
more » « less
Abstract We propose the use of a second surface mirror as a displacement plate-beamsplitter to provide significant simplification and cost reduction of time-of-flight anemometry (ToFA), without sacrificing precision and accuracy. These benefits are most pronounced for long-range applications. Our method’s principle benefits are due to the few and simple components it requires as well as low sensitivity to both temperature effects and light source incoherence. We found that precise and accurate results are possible using a common consumer mirror as the main optical element and an inexpensive diode laser as the light source, which could broaden access to laser anemometry and make many industry applications economically feasible. The nature of the design also permits an increase in range for a given laser power since the method can utilize the entire optical area of the focusing lens/mirror independent of other design considerations and the cost of a flat second-surface mirror is usually negligible. To characterize the performance of this method, we develop a Cramer–Rao bound (CRB) for a general class of ToFA’s with multiple Gaussian beams under signal-independent Gaussian white noise. For a given measurement volume, the lowest velocity uncertainty is achieved by creating a standard two-sheet geometry: power-matching the first two beams by adjusting the beamsplitter and blocking the rest of the beams is optimal. However, keeping the higher order beams permits determination of flow direction. Conditions to achieve beam power-matching are given. An anemometer is built using a diode laser with 12 mw 405 nm beam using a total of just three transmitting optical components. Our setup has an accuracy of 99.1%. The worst-case precision of 96.7% nearly achieves the CRB, although optimizing the setup more can lower the bound, and therefore allow increase in the performance by an order of magnitude or more.
