The Laser Diode. Figure 1. Internal structure of a typical diode laser.

Transcription

1 The Laser Diode 1 Introduction This set of laboratory experiments is designed to have you become familiar with the properties of laser diodes and to become familiar with optical spectrum analyzers typically used in fiber and laser instrumentation to asses spectral characteristics of sources. In addition you will become familiar with the technique of using an optical grating to stabilize and to tune the frequency of a laser diode. The discussion and figures in this section are for the Hitachi 633 nm, HL6314MG diode laser. Figure 1 shows the diode laser with a cut-away of the housing where the chip mounts. Figure 2 shows the laser chip structure, typical chip dimensions, the direction of the forward current, and the radiation pattern. The radiation is produced in the active layer, which is a small fraction of the height of the chip, hence the radiation is diffracted when it emerges from the active region analogously to the diffraction of radiation passing through a narrow slit. The diffraction produces the radiation pattern shown in Fig. 2. The polarization of the emitted light is in the plane of the active junction, and therefore parallel to the short axis of the elliptically patterned light field. Figure 1. Internal structure of a typical diode laser. Figure 2. Chip structure of a typical diode laser. Figure 3 shows the package type and the internal circuit of HL6314MG. Notice that the package includes a built-in photodiode for the power-monitoring purpose. The common pin of the laser diode and the photodiode is usually designated as the ground pin that is connected to the case. Figure 4 shows the actual package dimensions and the pin layout. These two figures allow you to determine the Diode Laser Experiment Page 1 of 10

2 appropriate mechanical mounting configuration and electric connection for the laser diode. Figure 5 displays a number of important operating characteristics of the HL6314MG laser diode. The panel in the upper left corner shows the optical output power versus the forward bias current, measured under various temperatures. You should notice two important things. First, the laser diode output power rises sharply after the forward current exceeds a certain value, namely the lasing threshold current (25 ma). Second, the value of the threshold current increases with the rising temperature of the laser diode. The maximum operating forward current (50 ma) and power output (3 mw) are specified by the manufacturer for each diode laser. The upper right corner in Fig. 5 shows the relationship between the output power of the laser diode and the current of the monitoring photodiode (under a reverse bias of 5V). A calibrated external power meter would allow you to verify this relationship. The ellipticity of the emitted light field pattern is shown in the left lower corner of Fig. 5. The right lower corner of Fig. 5 shows that the laser diode operates in an increasingly coherent fashion when its output power is increased, with a more effective suppression of the side-modes and a narrower linewidth of the main mode. Figure 3. Package type and internal circuit of HL6314MG laser diode. Figure 4. Package dimensions of HL6314MG laser diode. Diode Laser Experiment Page 2 of 10

3 Figure 5. Important laser diode operating characteristics. The case temperature of the diode laser plays an important role in laser operations and Figure 6 shows the various effects. In general, when the case temperature rises, the diode laser operates less efficiently, with a rising threshold and decreasing slope efficiency (a conversion factor from the injection current to the actual optical output power). These scenarios are clearly shown in the top two panels in Fig. 6. Case temperature can be used to tune the operating wavelength of the laser diode. However, the wavelength tuning curve is not smooth over a large range of temperature changes, rather it will display a staircase type of tuning characteristics, as shown in the right lower corner of Fig. 6. This is because the laser diode actually jumps to different longitudinal modes when the temperature changes over a large range. While it is possible to change the laser wavelength by a couple of nanometers when the case temperature changes by 10 degree for example, the temperature tuning coefficient for a particular longitudinal lasing mode (without mode-hopping) is usually much smaller. The capability of continuous tuning of the diode laser wavelength without mode-hop is important for a number of applications in the AMO physics, including precision spectroscopy, laser cooling, and chemical sensing, etc. Combined adjustment of temperature and injection current can accomplish wavelength tuning to a certain degree, but usually within a fairly limited range. External cavity laser Diode Laser Experiment Page 3 of 10

4 diodes help solve this problem. Ordinarily, for a solitary laser diode, the laser cavity is formed between the two cleaved facets of the semiconductor chip. The facets are smoother and flatter than any mechanically polished mirror. If there are no coatings on the end surfaces of the laser chip, then the reflectivity R of a surface is given by: R = n n 2 c a, where n c and n a, are the indices of refraction of n c + n a the chip and air, respectively. Sometimes a diode laser manufacturer provides a reduced reflective coating on the output facet of the chip; therefore, the reflectivity of this facet is less than that calculated in Exercise 1. When the output of the diode laser is being returned with the first-order diffracted beam from an external grating, an external laser cavity is formed between the grating and the far chip facet. This would be particularly true if the grating diffraction efficiency (40% or more) dominates over the residual reflectivity of the front facet and the external cavity laser will operate more reliably and stably. The laser wavelength can now be tuned by displacing and/or rotating the grating. A carefully designed external cavity diode laser can pull the wavelength of the original laser diode by more than 10 nm without adjusting the injection current or chip temperature. It can also tune a single longitudinal laser mode by tens or even hundreds of GHz without mode-hop. Figure 6. The effects of case temperature on laser diode operations. Diode Laser Experiment Page 4 of 10

5 2 Experiment 2.1 PREPARATION You will be using a Hitachi laser diode from ThorLabs No. HL6316G. Begin by looking over the specifications sheet for the diode. In particular you will want to notice the operating current. You would normally also note the maximum allowed operating current, but this specification sheet does not provide it. Instead note the Typical Characteristics Curves where they show up to about 50 ma. The typical operating current is about 30 ma. Please do not exceed the maximal output power of 3 mw. Carefully handle the laser diode and avoid any static electric charge that can easily damage the diode. The diode needs to be mounted onto a TCLDM9 ThorLabs laser diode head with correct polarity. When you are ready to begin turning on equipment, first find the ThorLabs LDC 500 laser diode controller. Briefly read through the instruction manual. When you first turn on the power it enters a safe mode. Check that the LED for the correct polarity is lit, otherwise make a change in the polarity switch located at the rear panel of the controller. At this point you should also notice that the OPEN CKT LED should be off, indicating the laser diode has been connected to the controller. Set the Display indicator to show the current limit (ILIM). Located a small hole labeled LIM I. There is a small trimmer potentiometer that can be adjusted with a jeweler s screwdriver. You must set this trimmer for the maximum output current. The digital LED indicator will indicate the limit level in milliamps. Set it to be between 40 and 45 ma. Check to see that the mode indicator is on I for current. If not, press it once and the mode should change from P or power to I. When the mode switch is set to current, the diode laser driver will control the current to the diode so that it remains at a fixed level determined by the level knob. When the mode switch is set to power, the driver controls the laser diode to provide fixed output power. For the entire experiment leave the mode switch set to current. Turn the level knob all the way counter-clockwise. Now set the display indicator to ILD. As the level knob is increased in the clockwise direction the current indicator will increase. At this time no current is actually going to the laser, it only indicates the current that will be supplied when the laser is enabled. The laser will be turned on after a few other items are attended to. 2.2 THE TEMPERATURE CONTROLLER. Make a cable connection between the temperature controller (ThorLabs TEC 2000) and the laser diode head TCLDM9. Read the TEC 2000 instruction manual briefly and then turn on the temperature controller. As we now know, a stable case temperature is critical for reliable operations of the laser diode. In a research laboratory, a temperature controller for a laser diode is normally kept active all the time. The temperature controller consists of three parts, a temperature sensing unit, a feedback control loop, and a servo transducer for heating or cooling. The temperature sensor can be a standard thermistor or an IC temperature transducer such as an AD590. Both types of temperature sensing elements are included in the ThorLabs diode mount TCLDM9, one to provide a temperature error signal to feed into the feedback loop located inside the temperature controller, the other to provide an independent readout of the diode's temperature. The temperature controller provides current to the thermoelectric cooler, also built in with TCLDM9, to activate temperature corrections. When TEC 2000 is first turned on, the front panel LCD display should become active and you can display the desired measurement value, such as the actual temperature (Tact), the set temperature (Tset), the current limit to TEC (Ilim), and the actual TEC current (Itec). Select a suitable current limit Diode Laser Experiment Page 5 of 10

6 (Ilim) using the small trimmer potentiometer, for example, at 1.5 A. Toggle the temperature sensorsetting switch located at the rear panel and observe the front panel display. The OPEN CKT LED should be off. The temperature of the laser diode is controlled by the set temperature. Set the display mode to Tset and use the big adjustment knob to set Tset at about 25 degrees C. Turn on the On/off switch for temperature control loop, the associated LED ( Enable ) should turn on. Set the display mode to Tact and monitor how the actual temperature approaches the set value. Connect a BNC cable between the CTL OUT BNC Jack located at the rear panel and a digital scope. This output port allows additional monitoring of the temperature settling behavior on the digital scope. You need to experiment with the feedback control loop. Read in the TEC 2000 manual (page 16) on the procedure to adjust the Proportional (P), Integrator (I), and Derivative (D) gains. It will be good if you have some rudimentary feedback control understandings to appreciate the adjustment procedure to be performed here. Monitor the temperature-settling curve on the digital scope. You may need to adjust the time base on the scope to accommodate a global view of the settling behavior. Adjust the PID gains until you achieve an optimized compromise between the settling time and the number of error signal overshoots. Of course to record a new settling curve you will need to set a new temperature value, say 2 degrees away from the previous one. Please sketch several of these settling curves on your lab book and record the important time scales (setting time, number and size of overshoots, etc.). At this time, you can also turn on the signal generator, the Ando spectrum analyzer, the external photodetector or optical power-meter, and the PZT driver, etc. 2.3 LASER DIODE Now preset the laser diode current to about 15 ma on LDC 500 and push the Enable button to activate the injection current to the laser diode. Turn up the level knob so that the current goes up to about 30 ma. Notice the output from the laser (*DO NOT LOOK DIRECTLY INTO THE LASER BEAM!*) The laser should have a collimating lens in front of it, and nothing else. If there is other optics in the way, such as grating, move it so that you can see the laser output on the wall of the room. Adjust the collimating lens to obtain the best possible collimation judged by your eyes. This is an important step to achieve the best external cavity alignment later on. Using the photodetector module or the optical power meter you will measure the power output as a function of current and temperature. Connect a BNC cable from the photodiode output to the oscilloscope. Place the photodetector in front of the laser beam, note the signal on the oscilloscope and optimize the oscilloscope settings. Distinguish the laser power from the ambient room light. With the diode laser case temperature set at 25 degree C, make a measurement of the detected output power (proportional to the voltage on the oscilloscope) as a function of laser diode current. Start considerably below threshold. Then increase current in steps of 3-5 ma to the maximum current. Then go back and fill in more data points near the threshold value of current. Make sure you keep track of the oscilloscope voltage setting as you take measurements. Now retake the power versus current curve at two more temperatures. Take one at nearly the coldest temperature that the controller can manage (try 5 or 10 degree C). Then take another at about 40 degrees C. Finally, set the current to about 20-25% above the threshold value for 25 degrees C and take a power versus temperature curve. Do not exceed about 45 degrees C. (The specification sheet says 50, Diode Laser Experiment Page 6 of 10

7 but you must account for the fact that the thermistor does not really measure the diode temperature but the temperature of the diode mount. A much more efficient way to take the threshold curve is to use an external modulation input to the current drive. Set the temperature at 25 degree C and the current at 25 ma (or the threshold of your particular laser diode). Set the waveform on the external signal generator to be Triangle or Sawtooth ramp, with a frequency of about 5 Hz. Verify on the scope that the output signal from the signal generator is symmetric around 0 V and the amplitude does not exceed 1 V. Connect a BNC cable between the signal generator and the MOD IN BNC jack on the rear panel of LDC 500. Connect another BNC cable from the CTL OUT of the LDC 500 to the x- channel of the scope. Connect the photodetector output to the y-channel of the scope. Set the scope in X-Y mode. What you will observe on the scope is a direct display of the laser diode output power versus current, similar to that shown in Fig. 5. Record the trace on your notebook and compare with what you have measured earlier. When you are done, make sure you first disconnect the external signal input to the current driver before you turn the laser diode current to zero. 2.4 THE OPTICAL SPECTRUM ANALYZER. In this next portion of the lab you will be taking the wavelength spectrum of the diode. The optical spectrum analyzer is at first intimidating but in the end rather user friendly. Familiarize yourself with the controls with the help of the manual. To obtain a spectrum you must inject the laser diode light into a multi-mode fiber. One end of the fiber is already mounted in an optical claw. Take the short (15mm) focal length lens and place it after the collimating lens of the laser diode. Use a post-it note to determine the position of the focal point. Then place the fiber at this position. The easiest way to align the fiber more precisely is to look directly at the output of the other end of the fiber. (Yes your eye will be safe at these power levels.) When the fiber is correctly positioned the light through the fiber is quite bright (you should stop looking into the beam before it gets too bright), and you can easily use the fiber to illuminate a nearby object. You can use the power meter to maximize the output power through the fiber. When the fiber is aligned, screw the output end into the spectrum analyzer. Note that the fiber is keyed, and correctly fits into the receptacle in only one orientation. The threaded collar is annoyingly difficult to screw into the spectrum analyzer but it can be done. Set the spectrum analyzer center wavelength to 635 nm and the span to 20 nm. Set the wavelength resolution to the highest value (0.1 nm or 0.5 nm depending upon units). Set the laser current above threshold. Set the vertical scale to be logarithmic. You should easily observe a peak in the spectrum with no substantial noise for +/- 5 nm or so around the peak. At this point center the peak and change the span to 10 nm. Adjust the vertical scale accordingly. Now observe the spectrum for a few values of current both above and below the threshold. Record any interesting features and phenomena in the spectrum as a function of power. Note in particular the sinusoid-like pattern in the "pedestal" near the peak. Measure the spacing (in nm) between successive peaks. What do you conclude about the length of this laser diode? Now measure the position of the peak as a function of temperature for a current about 20-30% above threshold. Also at 25 degrees C measure the wavelength of the central peak as a function of current. Record your findings. Diode Laser Experiment Page 7 of 10

8 2.5 EXTERNAL CAVITY LASER DIODE For this portion of the experiment you will employ an optical grating of 1800 lines per mm. The grating is placed on a mirror mount with a piezo mounted under the horizontal adjustment screw, which can be used to tilt the grating by minute amounts. Place the grating as close as possible to the collimating lens mount. Tilt the grating so that the first order diffracted beam is reflected back into the diode. The zeroth order is available for measurement. It is a bit tricky to ensure that the 1st order is indeed retro-reflected back into the laser. Follow this procedure: Place the photodetector on the zeroth order beam. A.C. couple the oscilloscope input and turn up the gain almost all the way. Center the trace on the screen. As you move the vertical and horizontal adjustment screws on the grating mount observe the scope trace. Tap lightly on the mount. When the beam is properly retro-reflected the output power will appear to be very sensitive to your tapping. Maximize the tapping noise observed on the scope with the two grating adjustment knobs. At this point the grating is adequately adjusted. You can also directly monitor the diode laser threshold curve on the digital scope, as described at the end of section 2.3. As you adjust the grating to achieve different amounts of optical feedback, you will notice threshold reduction and optical power enhancement. Now align the fiber again to obtain a spectrum of the laser diode. Adjust the temperature to 25 degrees and the current to about 15% above threshold. Hook up the piezo to the pzt drive. Turn the voltage fully counter-clockwise and turn on the power supply. Slowly turn up the piezo voltage and record its value from the pzt driver monitor. Note the peak on the spectrum analyzer. Does it shift? If so, by how much? You may want to decrease the wavelength span. Be careful, sometimes the laser mode-hops between two modes. If you see the spectrum change, pin down whether or not a mode hop is responsible for the change. We will now proceed to calibrate the piezo wavelength control using an optical interferometer. 2.6 UNBALANCED MICHELSON INTERFEROMETER Remove the fiber from the beam path. Construct an unbalanced-path Michelson interferometer. Use Fig. 7 as a guide though your optical arrangement may look rather different in detail. Start with a path length difference as large as you can -- place an extra mirror in the longer arm if you wish to fold it to make it still larger. Record the distances from the beamsplitter to the respective end mirrors for later use. Hook up the piezo to the signal generator rather than the power supply. Send the output of the signal generator to the oscilloscope channel 2 as well. Connect the synch out output of the generator to the external trigger input of the scope. Set the sweep to triangular. Set the scope to external trigger. Put the output of the generator to about halfway. Observe the triangle wave on the oscilloscope. Diode Laser Experiment Page 8 of 10

9 L 1 M 1 Beamsplitter Photodiode Output signal L 2 Diode laser Grating M 2 Figure 7. Unbalanced Michelson interferometer for calibrating the piezo controlled grating tilt. Adjust the output amplitude and the dc offset of the signal generator so that the triangle output does not go negative, and so that the upper portion of the waveform does not distort. Set the signal generator frequency to between 40 and 100 Hz. Place the photodetector at the output of the Michelson interferometer and set the oscilloscope to view the photodetector and triangle waveforms simultaneously. You should see a sinusoid output from the detector that is synchronized with the triangle wave. If the laser is oscillating in more than one mode then the photodetector may show a beat envelop on the sinusoid. Try changing the laser current a bit if you run into this. If you can t get rid of it, then tolerate it. Your goal is simply to measure the change in piezo voltage that causes one full fringe at the photodetector. One full fringe corresponds to c/2(l2-l1) change in frequency. Observe the geometry used to tilt the grating with the small green piezo attached to the optical mount. Using your above measurement, the grating specs of 1800 lines per mm, and the fact that it is used in littrow configuration, estimate the displacement of the piezo in microns/volt. Congratulations! You re done! 3 Pre-lab Questions 1. The index of refraction of GaAs and air are about 3.6 and l.0. What is the reflectivity of the chip facet? Compare your answer with the reflectivity of the mirrors used in the HeNe laser, which is about Diode Laser Experiment Page 9 of 10

10 2. Assuming the length of the diode laser cavity is 250 µm and the index of refraction of the cavity is 3.6, show that the frequency spacing ν of the axial modes is 167 GHz. (You may want to refer to the discussion of longitudinal modes in the mode matching experiment.) 3. What does the axial mode spacing correspond to in wavelength spacing? (Give your answer in nm) 4. You will be working with a grating with 1200 groves per millimeter. Calculate the littrow, or blaze, condition for 635 nm. That is, calculate the angle of incidence of the laser beam such that the first order is retro-reflected. 5. The grating above is tilted by 1x10-3 rad. What is the change in wavelength, in nm, of the laser diode? 6. You are using an unbalanced Michelson with a path length difference of 1.12 m. You observe that 8.3 V on the piezo that tilts the grating causes one complete fringe to appear on the oscilloscope (that is, for example, the photodetector output changes by one peak to the next). What is the frequency tuning constant in MHz/Volt of the piezo? Diode Laser Experiment Page 10 of 10