An international comparison of eight I -stabilized semiconductor laser systems (DLs) has been carried out. Five of the DLs were extended-cavity lasers (ECLs) using extra-cavity saturation spectroscopy; another was a microlens-mounted diode modified to have weak optical feedback, stabilized using the same technique; the seventh ECL was stabilized using frequency-modulated spectroscopy. The final DL was a simple laser diode at 635 nm locked with a digital system on a linear absorption of iodine. The P(33) 6-3 transition of iodine was first used to compare the first seven DLs with a He-Ne laser stabilized on the R(127) 11-5 transition of iodine. The relative frequency stability of these lasers was between 5 parts in and 7 parts in for a sampling time of 1 s, with the best results less than 2 parts in over 1000 s. The frequency repeatability measured during one week was of the order of a few tens of kilohertz. This large fluctuation was caused by poor adjustment of the electronic offset of two of the lasers. For the well-corrected lasers, the repeatability was within a few kilohertz. A study of stabilization on the strong absorption group of transitions R(60) 8-4, R(125) 9-4 and P(54) 8-4, located about -12 GHz from the R(127) 11-5 transition, was also carried out. For the first time, a short-term frequency stability better than that of the classical He-Ne laser around 633 nm has been achieved with a relative frequency stability of 4 parts in for 1 s.

No abstract

A comparison of three 127 I 2 -stabilized extended-cavity diode lasers has been carried out between the Danish Institute of Fundamental Metrology (DFM) and the Bureau International des Poids et Mesures (BIPM). Two of the lasers are stabilized using intra-cavity saturation spectroscopy, the third one using extra-cavity saturation spectroscopy. All lasers are locked using the third harmonic detection technique on different hyperfine components of the P(33)6-3 transition. The range of frequency differences measured during the four days of the comparison was from -340 kHz to 150 kHz. The large fluctuations are due to the poor repeatability of the extended-cavity diode lasers with intra-cavity absorption cells (ECL4 and BIREL1-1). The frequency stability and repeatability of the diode laser (ECL1) using an external absorption cell is roughly one order of magnitude lower in performance than the He-Ne lasers. Using an iodine-stabilized He-Ne laser (DK3) as reference and the recommended values given in the practical realization of the definition of the metre, we found a mean offset from the expected frequency difference between ECL1 and DK3 of -12 kHz and a standard uncertainty of 10 kHz.

Results from an international comparison of 12712 He-Ne stabilised lasers at 633 nm wavelength of the Institute of Atomic Physics-IFTAR, Department of Lasers (IFA-IFTAR), participating with the laser RO. 1 and of the Bureau International des Poids et Mesures (BIPM), participating with the laser BIPM7 are reported. The frequency difference determined from the last three measurements resulted as v = JPM7 -"i.o. i = ( 38.2 1.3 ) kHz. Parameterswhich affected the reproducibility of the frequency for the lasers such as modulation width, iodine pressure (temperature) and laser intracavity power were also investigated. The frequency stability, expressed as the relative Allan variance, was measured for sampling times in the interval from 1 s to 1000 s and its value for a sampling time of 100 s resulted as 3.25x1012.

No abstract