Two-photon spectroscopy of rubidium using a grating-feedback diode laser

 
Zoom inZoom inZoom inZoom in
Zoom outZoom outZoom outZoom out
Go homeGo homeGo homeGo home
Toggle full pageToggle full pageToggle full pageToggle full page
Rotate leftRotate leftRotate leftRotate left
Rotate rightRotate rightRotate rightRotate right
Two-photon spectroscopy of rubidium using a grating-feedback diode laser Abraham J. Olson, Evan J. Carlson, and Shannon K. Mayer Department of Chemistry and Physics, University of Portland, Portland, Oregon 97203 (Received 29 July 2005; accepted 13 January 2006) We describe an experiment for investigating the 5S|/2— >5D5/2 two-photon transition in rubidium using a single grating-feedback diode laser operating at 778.1 nm (385 THz). Continuous tuning of the laser frequency over 4 GHz allows for the clear resolution of the Doppler-free spectral features and allows accurate measurement of the hyperfme ground-state splitting. A direct comparison between Doppler-broadened and Doppler-free spectral features is possible because both are distinctly evident in the two-photon spectra. By independently modifying the polarization state of the two laser fields, the impact of electric dipole selection rules on the two-photon transition spectra is investigated. This experiment is a valuable addition to the advanced undergraduate laboratory because it uses much of the same equipment as the single-photon saturated absorption spectroscopy experiment performed on the 5S1/2 — >5Py2 transition in rubidium (A. =780. 24 nm) and provides students with an opportunity to investigate characteristics of atomic spectra not evident in the single-photon experiment. © 2006 American Association of Physics Teachers. [DOI: 10.1119/1.2173278] I. INTRODUCTION Tunable diode lasers are used extensively in atomic phys¬ ics research. In recent years grating-feedback diode laser sys¬ tems have been developed that are suitable for construction by undergraduates.1-4 These narrow-band (<1 MHz) laser sources provide several milliwatts of optical power (10-80 mW for the laser systems in Refs. 1-4) and have a typical frequency scanning range of 5-10 GHz. Several atomic physics experiments have been developed for the advanced undergraduate laboratory to acquaint stu¬ dents with the basic techniques of laser spectroscopy and to investigate atomic properties. Recent atomic physics experi¬ ments that employ tunable diode lasers include the laser spectroscopy of rubidium and cesium,1 the temperature de¬ pendence of Doppler-broadening in the absorption spectrum of rubidium,5 atomic hyperfine structure studies in atoms,6 observation of the Faraday effect using the D2 resonance line in a rubidium vapor,7 and the laser spectroscopy of lithium8 and the cesium dimer.4 Doppler-free saturated absorption spectroscopy of ru¬ bidium is among the simplest experiments using tunable di¬ ode lasers and the experiment is now widely used in the upper-division physics laboratory. The single-photon spec¬ troscopy experiment employs a tunable diode laser tuned to the 5^1/2 — » 57*3/2 transition (\ = 780.24 nm) in rubidium. Saturated absorption spectroscopy provides a valuable op¬ portunity to observe the effect of Doppler broadening in atomic spectra and to investigate laser-saturation techniques for extracting the hyperfme structure features of the atomic system. A thorough description of saturated absorption spec¬ troscopy can be found in Refs. 1 and 10. Reference 11 also reviews the relevant theory of Doppler-free saturated absorp¬ tion spectroscopy and provides problems suitable for undergraduates.14 An investigation of the 5Sl/2—>5D5/2 two-photon transi¬ tion in rubidium (\ = 778.1 nm) is a valuable complement to the single-photon spectroscopy experiment because it uses much of the same equipment, is conveniently accessible to a grating-feedback diode laser system designed for single- http://aapt.org/ajp photon spectroscopy at 780 nm, and offers the opportunity to directly resolve the hyperfine levels of rubidium without the presence of Doppler broadening. In addition, the experiment provides an opportunity for the direct investigation of the relation between laser light polarization, quantum mechani¬ cal selection rules, and the observed atomic spectra. Two-photon transitions have historically been studied us¬ ing stabilized tunable dye lasers and titanium sapphire lasers.12 Subsequent observation of two-photon transitions in rubidium using AlGaAs diode lasers13 14 make these experi¬ ments more accessible to undergraduates. Two-photon tran¬ sitions in rubidium are of particular interest as new optical frequency standards due to their transition wavelength and narrow linewidth (~300 kHz linewidth for the two-photon transition).12 Optical frequency references have fiindamental and commercial applications, including uses in metrology, precision measurements, and the determination of fundamen¬ tal constants. Precise frequency references in the near- infrared make enhanced studies of atomic spectra possible. Commercially, this rubidium two-photon transition is useful for optical communications work. Frequency stabilization of a frequency-doubled 1.55 /xm laser to the two-photon ru¬ bidium line provides a valuable frequency reference for the I.5 yum optical communications band.15,1 In this paper we describe the relevant theory for two- photon spectroscopy. We provide a detailed description of the experimental procedure and equipment, including refer¬ ences for commercial sources of equipment17 and alterna¬ tives suitable for construction by undergraduates. We also give some typical experimental results and calculations that can be incorporated into the laboratory. II. THEORY An energy level diagram of the relevant transitions in ru¬ bidium is shown in Fig. 1 . The brightest spectral line, corre¬ sponding to the 5Sy2—>5P2/2 D2 line, has a transition wave¬ length of 780.24 nm (vacuum wavelength) and is often used for single-photon saturated absorption spectroscopy. The two-photon transition we investigated utilizes the 5 D5/2 ex- © 2(106 American Association of Physics Teachers 218 Am. J. Phys. 74 (3). March 2006 218