A Narrow-Band Tunable Diode Laser System with Grating Feedback S.P. Spirydovich Draft Abstract The description of diode laser was presented. The tuning laser system was built and aligned. The free run beam wavelength as well as diffracted beams wavelength was measured and tuning was applied. The free-run spectrum graph was presented. Introduction Among many other advantageous features of diode lasers there are their small size, high efficiency, low price, simple pumping requirements, reliability to be a source of narrow-band (<1MHz) light, power, wavelength coverage and the possibility to modulate their injection current at high frequencies. Diode lasers can emit in a continuous or in a pulsating way, but in both cases the maximum emitted power is usually lower than a few tens of mw. Nevertheless for many applications it is siutable to have high power coherent emission available from small sources. The tunable diode lasers are often used in atomic physics, because sources of laser light can be tuned to particular atomic transition. The tuning means the change of a diode laser s wavelength. And the junction s temperature and current density determine this wavelength. Among some techniques for controlling and narrowing laser output spectra (tuning) we choose the optical feedback method. According to this method the linewidth will be reduced if the quality factor Q of laser s resonator is increased. For the optical element such a feedback one can use gratings, elatons, fiber cavities, phase conjugate or simple mirrors. The main negative part of the method of optical feedback is that laser s frequency has to be stabilized. To obtain stable frequencies one should reduce such major problems as vibrations, thermal changes and air currents. Previous experimental results [1] showed that even if laser s output is bigger than 1MHz, by applying continuous tuning it is possible to have laser light to be confined only over some wavelength region. The system for our experiment was already previously used in experiments in optical cooling and trapping [1]. Our usage of similar equipment is eventually to repeat and improve experiment of laser spectroscopy of rubidium [2]. Experiment As shown in (fig. 1) there are three essential elements, a diode laser, a collimating lens and a diffraction grading. The system with AlGaAs laser in the near infrared region of spectrum produces more than 10mW of light with bandwidth of well under 1MHz.
Figure 1. Diode Laser System with Grating Feedback To run laser the low-noise current source is needed. For this purpose a 9V battery was chosen as the power supply. We tried to mount the optical part of the system in such a way so that lens provides proper collimating relative to laser. The laser beam reflects the diffraction grating and returns to laser as the diffracted in the first. The possibility of rapid modulation of amplitude and frequency by changing the injection current is one of the very important advantages of diode lasers.
To change the cavity length the system has piezoelectric (PZT) disks (fig.2) and to stabilize temperature there is an aluminum enclosure. Figure 2. Piezoelectric disks We should say that necessary step is to have temperature control, since according to [1] the dependence of temperature is fairly significant (fig.3).
Figure 3. Laser output temperature vs. thetemperature of the laser case Our steps in preparation of the system were the following. First we cleaned the grating with alcohol in ultrasound. Second we assembled the system primarily only two main parts; diode laser and grating. We measured laser beam wavelength at without and with tuning. Second group did latter one while they were observing track in rubidium vapor, which should happen only at wavelength of 780nm. The free-run laser s wavelength was measured by spectrometer at 782.76 nm with width of 0.05nm. (fig.4). To estimate wavelengths of two observed diffracted beams we used monochrometer, which had not as good accuracy as spectrometer, so roughly each beam had wavelength 781±1nm.
To observe the saturated absorption spectra of rubidium the wavelength of laser light should be 780nm. Beam layout for saturated absorption is shown in (fig.5). At this stage we could not observe any track of absorption, because as it was mentioned above laser was not yet tuned appropriately. For the observation and measurement we applied some new techniques. First of all since wavelength of 780nm is invisible there were three possible ways available to observe light of laser beams: by means of IR-card, IR-viewer and CCD camera with TV. The last one was the easiest way to follow reflected beams. One more thing should be mentioned is reflection of light in the tube with rubidium vapor. We could not observe any track if there were even insignificant light sources other than laser. So the most appropriate place to set up experiment and take measurements was dark room. For the index guided lasers such as AlGaAs linewidths of 20 to 50 MHz are typical. In our case diode laser can be tuned over a wavelength of about 20 nm. We also have laser frequency changed due to temperature (fig. 3), where a temperature-tuning curve of an ideal device is a stair case with sloping steps. Since we have tuning wavelength interval of 20 nm, we need to cool laser, so we could approach region of 780nm. Future Results By the time when this paper is being written another group did the second part of this experiment. Their results confirmed the existence of absorption spectra of rubidium vapor. The importance of temperature control was taken into account, so they have had stabilized temperature of diode as well as baseplate. It is necessary to mention that the third group developed cooling system to stabilize thermal changes of laser. The explanation of technical specifications could be found in their paper. Conclusion We observed track in rubidium vapor at wavelength of 780nm, which means that laser system was tuned and finally we can have high-power coherent emission available from fairly small laser. Acknowledgments I would like to thank Prof. S. Durbin, E. Dedrick and A. Soliman for their discussions and patience. References
1 Using Diode Lasers in Atomic Physics C.E.Wieman and L. Hollberg Rev. Sci. Instrum. 62 (1), January 1991 2 K.B. MacAdam, A. Steinbach and C. Wieman Am. J. Phys. 060 (12), December 1992
Intensity vs. Wavelenth Diode L 782.5 782.6 782.7 782.8 782.9 783 783.1 Wavelength (nm) Figure 4. The free run laser beam power spectrum
Figure 5. Beam layout for saturated absorption