Open Access
Add to BibSonomyAbstract
We report a monolithic gain-switched single-frequency Yb-doped fiber laser with widely tunable repetition rate. The single-frequency laser operation is realized by using an Yb-doped distributed Bragg reflection (DBR) fiber cavity, which is pumped by a commercial-available laser diode (LD) at 974 nm. The LD is electronically modulated by the driving current and the diode output contains both continuous wave (CW) and pulsed components. The CW component is set just below the threshold of the single-frequency fiber laser for reducing the requirement of the pump pulse energy. Above the threshold, the gain-switched oscillation is trigged by the pulsed component of the diode. Single-frequency pulsed laser output is achieved at 1.063 μm with a pulse duration of ~150 ns and a linewidth of 14 MHz. The repetition rate of the laser output can be tuned between 10 kHz and 400 kHz by tuning the electronic trigger signal. This kind of lasers shows potential for the applications in the area of coherent LIDAR etc.
© 2016 Optical Society of America
1. Introduction
Pulsed single-frequency fiber laser sources have attracted considerable attentions, because of their potential applications in various areas [1–3], such as acoustic phonons detection [1], terahertz wave generation [2] and coherent LIDAR (LIght Detection And Ranging) [3]. For example, a coherent LIDAR generally requires a single-frequency pulsed laser to generate a heterodyne signal for highly sensitive Doppler ranging sensing. The detection range requires one optical pulse in transit at one time. Therefore a specific repetition rate determines the detection range and the system sampling rate [3]. For adapting to the variable detection distance of the coherent LIDAR, a repetition-rate tunable single-frequency pulsed fiber laser is highly demanded.
One typical approach to achieve pulsed single-frequency laser output is to modulate CW single-frequency laser externally by using of acousto-optical or electro-optical modulators [1,3]. This method will not significantly impact the linewidth of the single frequency. However, these lasers have to use some additional active modulation components. The low duty ratio of the modulation makes the output power very low. Hence, the system usually need an additional amplifier. All of these will increase size, cost, and complexity. S. Jiang et al. reported another approach to obtain an all-fiber single-frequency pulsed laser [4, 5]. The pulsed oscillation was achieved by applying mechanical force on the cavity and modulating the fiber birefringence using a piezoelectric modulator. However, the mechanical force applied by the piezoelectric modulator will damage the fiber in a long-term operation.
On the other hand, an alternative approach of gain-switching is advantageous for achieving pulsed single-frequency output in a simple all-fiber monolithic configuration requiring no additional modulation components inside or outside the cavity. J. Geng et al. reported the first single-frequency gain-switched fiber laser [6]. A gain-switched Ho-doped fiber laser was obtained via in-band pumping, using a Q-switched Tm-doped fiber laser at 1.95 μm as the pumping source, since the Q-switched Tm-doped fiber laser had a nearly fixed repetition rate, the tunable repetition rate of the gain-switched Ho-doped fiber laser was not realized. Meanwhile the pump needs relatively high pulse energy (PE). All of these factors limited the application of this technology and made the system complex and expensive.
In this work, we demonstrate a monolithic, repetition-rate widely tunable, gain-switched single-frequency Yb-doped fiber laser. The single-frequency output is obtained by the DBR laser cavity. Similar to the hybrid pump scheme [7, 8], the modulated LD pump at 974 nm contains both CW and pulsed components. To reduce the requirement of the pulse pump power, the CW component is set slightly below the threshold of the single-frequency fiber laser. The gain-switched oscillation is then manipulated by the pulsed component in the pump. The repetition rate of the pulsed output can also be tuned by controlling the repetition rate of the pulsed component of the pump. The repetition rate of the laser output can be tuned between 10 and 400 kHz. The tuning range is only limited by the power and modulated pulse width of the pump LD. The pulse duration and the linewidth are measured to be 150 ns and 14 MHz at the repetition rate of 100 kHz, respectively. The amplitude fluctuation of the pulse is observed to be less than 5% under the repetition rate of 100 kHz.
2. Experimental setup
The experimental setup of the gain-switched single-frequency Yb-doped fiber laser is shown in Fig. 1. A 974 nm LD pump is mounted on a diode driver (Thorlabs, CLD1015), which can drive the LD in CW mode and pulsed modes simultaneously. An electronic trigger activates the diode driver and the 974 nm pump diodes deliver pump pulses (300-800 ns) from 10 to 400 kHz. The trigger signal applied onto the LD driver is shown as the yellow waveform in Fig. 1. The trigger function can be expressed as
in which f(t) is a square wave signal with a high level of 10 volts and a low level of 0 volt respectively, and c is a constant, which is adjusted based on the threshold of the DBR fiber laser. The repetition rate of f(t) function can be tuned between 10 kHz and 400 kHz.
Fig. 1 Schematic of the gain-switched single-frequency Yb-doped fiber laser. A 974 nm LD pump is modulated by a diode driver, which is triggered by the electronic square wave signal as shown the yellow waveform. The modulated pump is inputted in the DBR laser through the WDM coupler. The gain-switched laser is separated from the pump by the WDM coupler and output through the isolator.
The Yb-doped fiber DBR laser cavity is constructed in house [9]. A commercial Yb3+-doped fiber (Coractive, DCF-YB-7/128-FA) with an absorption coefficient of ~18 dB/cm at 974 nm was employed as the active gain medium. The fiber with the length of 1cm offers sufficient gain. The facet ends of a 1-cm-long active fiber are directly fusion-spliced with a narrow-band fiber Bragg grating (FBG) with 80% reflectivity and a broad-band FBG with 99% reflectivity, respectively. There is no other passive fiber in the cavity. The equivalent cavity length is about 2 cm in consideration of the effective length of the grating region, corresponding to the FSR is 5 GHz. There is only a few longitudinal modes permitted in the cavity, because of the bandwidth of the high reflectivity and the low reflectivity FBG is 0.25 nm and 0.07nm, respectively. The single-frequency operation can be realized by lasing of only one longitudinal mode through gain competition. The end of the FBG with high reflectivity is cut with an angle of 8°for the purpose of antireflection. The DBR laser is connected to the LD by a wavelength division multiplexing (WDM) coupler, and then backward pumped for separating the output and the pump. The signal port of the WDM coupler is spliced with an isolator to avoid the interference of the light echo.
3. Result and discussion
3.1 Pulse duration
The output pulse characteristics can be influenced by many factors, such as the pump PE, the cavity length, the pump absorption coefficient, the pump pulse repetition rate, etc. The pulse duration of the gain-switched laser output depends on the absorbed pump power (Pabs), the cavity length (L), the fiber core area (A), the pump photon energy (hvp) and the emission cross section (σemi) at the output wavelength [10–13], by approximating laser pulse duration (tLp) as being proportional to the period of relaxation oscillations. The relation can be shown as:
Where c is the speed of light in the medium and the cavity length (L) is the sum of the passive and the active fiber length L = Lpas + LYb. The absorbed pump power Pabs can be deduced from the incident pump power Ppump, the pump absorption coefficient αdB and the active fiber length LYb:
Equation (2) indicates that the laser pulse duration tLp is proportional to the square root of the cavity length and inversely proportional to the square root of the absorbed pump power. More universally, Eq. (2) can be expressed as
where K is the constant including the information of the emission cross-section and the pump wavelength, and the variable of X represents the term of (AL/Pabs)1/2indicating the absorbed pump intensity per fiber length, respectively. Taking all these related parameters i.e., the pump wavelength, the emission cross section, the cavity length, the pump power and the core area, into account, Fig. 2 summarizes the relation between the laser pulse duration tLp and the variable X of this work, in comparison with the data extracted from Ref [13]. and [14]. It is seen that there exists a good linear relation between tLP and X, regardless of the type of the rare-earth dopant ions, the pump and the lasing wavelengths, etc. The slope of K is linearly fitted to be 12.9 ± 0.3%. It must be noted that the deviation of the data points shown on Fig. 2 is believed to be due to the original approximation of Eq. (2) and the uncertainty of the information (e.g., the emission cross section, the cavity length, the core area, etc.) given from the Refs [13, 14]. All in all, as an empirical relation, Eq. (4) is a good guidance for estimating the pulse duration of the gain-switched lasers.Fig. 2 Empirical linear relation between pulse duration and X variable, based on the data from this work and other works [13, 14].
Due to the short cavity employed in our work, with low pump power, similar values of X can be obtained in comparison to the cases of using long cavity and high pump power when the output pulse duration is in the same range [13, 14]. Therefore the gain-switched single-frequency laser can be obtained in our work. Also, according to Fig. 2 and Eq. (4), it is expected that we can further reduce the pulse duration from the current ~150 ns down to ~50 ns by increasing absorbed pump power by about one order of magnitude, without changing other cavity conditions.
In order to reduce the requirement on the pump energy, the CW component of the pump is set to be 9.1 mW, which is just below the threshold of the Yb-doped fiber DBR laser. As shown in Fig. 3(a), the red dotted line indicates the threshold of the laser. Above the threshold, the gain-switched oscillation is then trigged by the pulsed component in the pump.
Fig. 3 (a) Temporal profiles of pump pulses and gain-switched pulses with different pump energy. The red dotted line indicates the threshold of the laser. (b) Comparison of pulse duration and build-up time under the different repetition rates. The inset shows the enlarged profile of the gain-switched output pulse.
Temporal profiles of the pump pulses and the gain-switched laser pulses under different pump parameters are measured by a fast photodiode (Thorlabs, Model DET08CFC with a bandwidth of 5 GHz) and a digital oscilloscope (as shown in Figs. 3(a) and 3(b)). The pulse duration, the energy, and the build-up time of the generated pulses depend on the pump energy and the repetition rate [11–13]. As shown in Fig. 3(a), under a fixed repetition rate of 100 kHz, when the pump PE increases from 58 nJ to 130 nJ, the pulse duration decreases from 240 ns to 150 ns. The build-up time drops with the increase of the PE too. In Fig. 3(b), one can see that the increase of the repetition rate also causes the decrease of the pulse duration and the build-up time, under the fixed pump PE of 130 nJ. The pulse duration is measured to be 180 ns, 150 ns, and 130 ns, corresponding to the repetition rate of 50 kHz, 100 kHz, and 200 kHz respectively. These results are in approximate agreement with the above Eq. (2). The slope efficiency of the output pulse to the pump pulse component is around 8%. From the inset of Fig. 3(b), one can see that the profile of the signal pulse is smooth, indicating that there is no beating between different frequencies and hence the gain-switched fiber laser works in the single-frequency state.
3.2 Pulse train and spectrum
As shown in Fig. 4(a), the corresponding output powers are 0.153 mW (10 kHz)、1.04 mW (100 kHz)、2.15mW (200 kHz)、3.32 mW (300 kHz)、4.72 mW (400 kHz), respectively. The peak-to-peak amplitude of the gain-switched pulse trains shows good stability, with no more than 5% fluctuation at the repetition rate of 100 kHz. With the increase of the repetition rate, the amplitude fluctuation becomes larger, but is still less than 10%. The reason may be because the CW component of the pump power is close to the threshold of the DBR laser and the output pulse is sensitive to the small fluctuation of the incident pump power. The fluctuation of the incident pump power is caused by the noise of the driver current, the temperature perturbation of the LD, the fiber splices, and so on. When the repetition rate increases, the average pump power increases and more heat will be generated, for example in the pump LD and the splice points. These will influence the stability of the pulse train. The large pump power and pump fluctuation will increase the frequency and intensity of the relaxation oscillations of the laser [7, 15]. These will reduce the stability of the pulse train in high frequency region.
Fig. 4 (a) Traces of pulse trains under different repetition rates. From top to bottom, the corresponding repetition rate is 10 kHz, 100 kHz, 200 kHz, 300 kHz and 400 kHz, respectively. (b) The spectrums of the gain-switched single-frequency laser with different pumping modes (Note that the OSA resolution is set at 0.02 nm). The black line is the spectrum of the pulsed output with the repetition rate of 100 kHz and the pulse width of 150ns under the hybrid pumping. The blue one is the ASE under the CW pumping.
The output spectrum of the gain-switched single-frequency laser is recorded by an optical spectrum analyzer (AQ6373, Yokogawa). Figure 4(b) shows the spectrums with different pumping modes. The black line is the spectrum of the pulsed output with the repetition rate of 100 kHz and the pulse width of 150 ns under the hybrid pumping. The blue one is the ASE under the CW pumping. It is seen that the optical signal-to-noise ratio (OSNR) of the pulsed output is more than 50 dB. The spectrum of the pulsed output exceeds the ASE more than 40 dB. The contrast is more than 40dB between the pulse and the background output.
3.3 Single-frequency operation
The single-frequency operation is confirmed by using a scanning Fabry–Perot interferometer (FPI) (Thorlabs, Model SA210-8B) with a 10 GHz free-spectrum range and a resolution of 67 MHz. As shown in Fig. 5(a), there is only one peak within the free-spectrum range of 10 GHz, thereby indicating the single-frequency laser operation. We use another scanning FPI (Thorlabs, Model SA200-8B) with a 1.5 GHz free-spectrum range and a resolution of 7.5 MHz to measure the linewidth. Figure 5(b) shows the peak spectrum of the scanning FPI with a 1.5 GHz free-spectrum range. Because of the pulsed output, the spectrum is discrete and the interval corresponds to the period of the gain-switched laser. The full width at half maximum (FWHM) of the envelope of the spectrum is 14 MHz. This is much larger than the transform-limited linewidth of 2.93 MHz, which is calculated by the limit of the Fourier transformation with a given pulse width of the gain-switched pulses of 150 ns. The reason of the linewidth broadening may come from the pump-induced refractive index change in active fiber [16, 17]. The refractive index change based on the pulsed pumping impacts the FSR of the laser cavity, and increases the longitudinal mode shift and the frequency noise. These will broaden the linewidth of the laser. Therefore, it is possible to further compress the pulse width to the transform-limited by using higher pump power.
Fig. 5 (a) Verification of single-frequency operation using a scanning FPI with FSR of 10 GHz. (b) The linewidth of the laser is about 14 MHz measured by the scanning FPI with a 1.5 GHz free-spectrum range and a resolution of 7.5 MHz. The discrete peak spectrum is caused by the pulsed output.
4. Conclusion
In summary, a monolithic all-fiber gain-switched single-frequency Yb-doped fiber laser with widely tunable repetition rate has been demonstrated. A modulated LD containing both CW and pulsed components in the output has been employed as the pump. In order to further reduce the requirement on the average pump power, the power of the CW component has been set close to the threshold of the single-frequency fiber laser and the gain-switched oscillation above the threshold is then mainly manipulated by the pulsed component in the pump. The repetition rate of the laser output can be tuned from 10 kHz to 400 kHz and the pulse duration is around 150 ns. An empirical relation has been obtained based on the experimental results from this work and referenced works on gain-switched fiber lasers and it can be expected that we can further reduce the pulse duration by increasing the pump power. The selected configuration of the laser cavity has shown high adaptability and stability. This truly monolithic all-fiber single-frequency laser configuration opens the way for many applications that require a laser source with narrow linewidth, high coherence and tunable repetition rate.
Funding
National Natural Science Foundation of China (NSFC) (61527822, 61235010, 61378088, 61307054).
References and links
1. P. Dainese, P. S. J. Russell, G. S. Wiederhecker, N. Joly, H. L. Fragnito, V. Laude, and A. Khelif, “Raman-like light scattering from acoustic phonons in photonic crystal fiber,” Opt. Express 14(9), 4141–4150 (2006). [CrossRef] [PubMed]
2. M. Leigh, W. Shi, J. Zong, Z. Yao, S. Jiang, and N. Peyghambarian, “Narrowband pulsed THz source using eyesafe region fiber lasers and a nonlinear crystal,” IEEE Photonics Technol. Lett. 21(1), 27–29 (2009). [CrossRef]
3. R. J. De Young and N. P. Barnes, “Profiling atmospheric water vapor using a fiber laser lidar system,” Appl. Opt. 49(4), 562–567 (2010). [CrossRef] [PubMed]
4. M. Leigh, W. Shi, J. Zong, J. Wang, S. Jiang, and N. Peyghambarian, “Compact, single-frequency all-fiber Q-switched laser at 1 microm,” Opt. Lett. 32(8), 897–899 (2007). [CrossRef] [PubMed]
5. J. Geng, Q. Wang, J. Smith, T. Luo, F. Amzajerdian, and S. Jiang, “All-fiber Q-switched single-frequency Tm-doped laser near 2 mum,” Opt. Lett. 34(23), 3713–3715 (2009). [CrossRef] [PubMed]
6. J. Geng, Q. Wang, T. Luo, B. Case, S. Jiang, F. Amzajerdian, and J. Yu, “Single-frequency gain-switched Ho-doped fiber laser,” Opt. Lett. 37(18), 3795–3797 (2012). [CrossRef] [PubMed]
7. J. Yang, Y. Tang, and J. Xu, “Development and applications of gain-switched fiber lasers [Invited],” Photon. Res. 1(1), 52–57 (2013). [CrossRef]
8. S. Yan, Y. Wang, Y. Zhou, N. Yang, Y. Li, Y. Tang, and J. Xu, “Developing high-power hybrid resonant gain-switched thulium fiber lasers,” Opt. Express 23(20), 25675–25687 (2015). [CrossRef] [PubMed]
9. Y. Hou, Q. Zhang, and P. Wang, “Frequency- and intensity-noise suppression in Yb3+-doped single-frequency fiber laser by a passive optical-feedback loop,” Opt. Express 24(12), 12991–12999 (2016). [CrossRef] [PubMed]
10. W. Koechner, Solid State Laser Engineering (Springer, 2006).
11. L. A. Zenteno, E. Snitzer, H. Po, R. Tumminelli, and F. Hakimi, “Gain switching of a Nd+3-doped fiber laser,” Opt. Lett. 14(13), 671–673 (1989). [CrossRef] [PubMed]
12. V. Agrež, R. Petkovšek, D. Sangla, J. Saby, R. B. Picard, and F. Salin, “Effect of repetition rate on gain-switched fiber laser output pulses,” Laser Phys. 24(10), 105108 (2014). [CrossRef]
13. C. Larsen, K. P. Hansen, K. E. Mattsson, and O. Bang, “The all-fiber cladding-pumped Yb-doped gain-switched laser,” Opt. Express 22(2), 1490–1499 (2014). [CrossRef] [PubMed]
14. S. Hollitt, N. Simakov, A. Hemming, J. Haub, and A. Carter, “A linearly polarised, pulsed Ho-doped fiber laser,” Opt. Express 20(15), 16285–16290 (2012). [CrossRef]
15. O. Svelto, Principles of Lasers (Plenum Press, 1998).
16. A. A. Fotiadi, O. L. Antipov, and P. Mégret, “Dynamics of pump-induced refractive index changes in single-mode Yb-doped optical fibers,” Opt. Express 16(17), 12658–12663 (2008). [CrossRef] [PubMed]
17. H. Tünnermann, J. Neumann, D. Kracht, and P. Weßels, “Gain dynamics and refractive index changes in fiber amplifiers: a frequency domain approach,” Opt. Express 20(12), 13539–13550 (2012). [CrossRef] [PubMed]
