Measuring & Understanding Pulse Characteristics of Femtosecond Fiber Laser - Prof. Jones, Lab Reports of Chemistry

Download Measuring & Understanding Pulse Characteristics of Femtosecond Fiber Laser - Prof. Jones and more Lab Reports Chemistry in PDF only on Docsity! Intensity Autocorrelation, F 2008. 1 R. J. Jones Optical Sciences OPTI 511L Fall 2008 Femtosecond fiber laser and pulse measurement: the intensity autocorrelation The generation of short optical pulses is important for a wide variety of applications, from time-resolved measurements utilizing the short pulse durations to nonlinear optics utilizing the high peak intensity of the pulses. The measurement and characterization of the pulse train emitted from a short pulse laser is critical in determining its performance, achievable peak powers, and the time resolution one has available for experiments. Below are some suggested measurements to characterize the fs fiber laser and better understand the instruments you will use in this lab. Using a beam splitter, you may want to observe some of these effects simultaneously (e.g. laser spectrum with OSA and output pulse train with fast photodiode and oscilloscope). (1) Observe the laser spectrum vs. injected current Due to the short length of the fiber laser output, please do not connect this directly to the OSA, but rather couple the light into the separate fiber provided. - At what current does the laser modelock? - Look for hysteresis in the modelocking behavior versus pump current. - Can you observe multiple pulsing? What is the spacing of the pulses? Can you see (resolve) these multiple pulses simultaneously on the photodiode? - What is the shortest pulse duration you can expect to measure with this laser? determine this by measuring the FWHM of the spectral distribution. - In the HeNe modelocking lab ~ 8-10 modes were lasing simultaneously. Approximately how many modes are lasing simultaneously in the present laser? Can you resolve them individually? (2) Observe the pulse train on fast photodiode vs. current - Observe the lasing threshold, modelocking threshold, and pump current at which the fiber laser emits multiple pulses. (3) Measure the signal (voltage) vs. applied laser current for the InGaAs compared to Si photodiodes. - Measure the slope of signal vs. current. For a linear photodiode, one would expect to see a fairly linear relation between average laser power and the applied current. The Si photodiode, with a larger bandgap, will be sensitive to the peak power coming from the laser, as the photocurrent requires absorption of 2 photons of the 1.55 micron laser light. - One difficulty with this measurement is finding a range over which the laser properties do not change, aside from the average power. For example, you will find a current at which the laser begins to lase, and a different current at which it begins to modelock. At higher currents yet the laser will emit multiple pulses. Intensity Autocorrelation, F 2008. 2 - You may also want to try different spot sizes on the Si photodiode to note any differences in its response. Ideally, for a fixed spot size, the signal will change quadratically with power. The smaller the spot size, the greater the signal response. This will be very different behavior than the InGaAs photodiode, which should responds linearly to the optical power. -Keep in mind the 2 photodiodes have very different surface areas, leading to very different response times (due to the capacitance of the device). The smaller InGaAs photodiode will be fast enough to see the pulse train directly, but not resolve their actual pulse duration. The large Si photodiode will have a much slower response time. (4) Intensity autocorrelation This is the primary measurement to make for this lab. The measurement consists of using the Michelson interferometer setup provided along with the 2-photon photodetector (Si photodiode). The delay can be scanned manually, or by driving the speaker with a ~24 Hz signal. Both are recommended. Make sure you can obtain a signal on the Si photodiode from each arm of the interferometer separately, then scan the delay and look for the pulse intensity autocorrelation. For slow, manual scans, you will be able to temporally resolve the interference fringes between the 2 arms simultaneously with the intensity autocorrelation. This is known as a “fringe resolved” or interferometric intensity autocorrelation. The signal to background should be a 8:1 ratio when properly aligned. (ie “signal” at long delays should be 1/8 of the signal at zero delay). See Figure 1 for examples of 2 pulses. Figure 1. Interferometric Intensity Autocorrelation. (See http://en.wikipedia.org/wiki/Optical_autocorrelation for full description) When scanning at 24 Hz (with a function generator driving the speaker), the slow Si photodiode response will not resolve these interference fringes, and they will be washed out. The function you will see is then the true intensity autocorrelation. Examples are given in Figure 2 below. In this case, a properly aligned instrument will yield a 3:1 ratio for the signal at zero delay vs. infinity. This autocorrelation signal is truly a convolution of the pulse profile with itself1. To determine the actual pulse duration, we can make a small approximation that the pulse shape is of the form sech2(t) (a good approximation for this type of fiber laser). In this case, the true pulse width can be determined by 1 The profile, by definition, should then also be symmetric. If not, the alignment must be wrong. What would the autocorrelation look like for 2 pulses?